{
  "title": "The Frontier Model Matrix",
  "description": "Context windows, output limits and pricing for the frontier LLMs an agent-builder reaches for (June 2026), enriched to full EAV depth per research/briefs/models.md. Claude rows are exact, verified against the Anthropic model catalog and AWS Bedrock model cards. Non-Anthropic rows list capability where it is publicly stable and defer pricing/context to the provider, because third-party prices move and we will not print a number we cannot vouch for. Every sourced value carries its primary source URL + last_verified; any value not confirmed against a primary source is a structured placeholder { value:null, verify_status:'verify-against-primary-at-build', source_hint:<primary URL> } rather than a guess.",
  "updated": "2026-06-15",
  "pricing_unit": "USD per 1M tokens (input / output)",
  "agent_readiness_score_formula": "round( 100 × ( 0.35·R_tool + 0.20·R_struct + 0.15·R_ctx + 0.20·R_cost + 0.10·R_latency ) ); each input normalized to [0,1]; R_ctx = min(effective_context/256000,1); tool-call reliability dominates (35). Full methodology on /models/agent-readiness-score. A model is scored only when every input is sourced; otherwise score_confidence: partial and agent_readiness_score is a structured placeholder.",
  "fields": [
    "id",
    "name",
    "vendor",
    "model_id",
    "context_window",
    "max_output",
    "input_per_mtok",
    "output_per_mtok",
    "strengths",
    "provider",
    "family",
    "release_date",
    "last_updated",
    "open_weights",
    "license",
    "params_total",
    "params_active",
    "tool_call",
    "reasoning",
    "structured_output",
    "attachment",
    "temperature",
    "knowledge_cutoff",
    "context_advertised",
    "context_effective",
    "price_input",
    "price_output",
    "price_cache_read",
    "price_cache_write",
    "cost_per_full_window",
    "cost_per_agent_task",
    "modalities",
    "gpqa_diamond",
    "swe_bench_verified",
    "terminal_bench",
    "tau2_bench",
    "bfcl_tool_use",
    "aa_index",
    "lmarena_elo",
    "tokens_per_sec",
    "ttft",
    "hallucination_rate",
    "agent_readiness_score",
    "score_confidence",
    "source_url",
    "source_type",
    "last_verified",
    "max_output_meta"
  ],
  "records": [
    {
      "id": "claude-fable-5",
      "name": "Claude Fable 5",
      "vendor": "Anthropic",
      "model_id": "claude-fable-5",
      "context_window": "1M",
      "max_output": "128K",
      "input_per_mtok": "$10.00",
      "output_per_mtok": "$50.00",
      "strengths": "Anthropic's most powerful, most intelligent model — a tier above Opus. Adaptive thinking; the model that built this site.",
      "provider": "Anthropic",
      "family": "Claude Fable",
      "release_date": {
        "value": "2026-06-09",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.anthropic.com/news/claude-fable-5-mythos-5 (Anthropic does not disclose parameter counts)"
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.anthropic.com/news/claude-fable-5-mythos-5 (Anthropic does not disclose parameter counts)"
      },
      "tool_call": {
        "value": true,
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15",
        "note": "Supported features at launch include the memory tool, code execution and programmatic tool calling."
      },
      "reasoning": {
        "value": "always",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15",
        "note": "Adaptive thinking is always on; thinking cannot be disabled. Extended thinking: No (overview table)."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/models/overview — schema-guaranteed structured-output flag not explicitly published for claude-fable-5 (no models.dev entry exists for this model)."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15",
        "note": "Vision listed in supported features; audio/video not listed."
      },
      "temperature": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/models/overview — temperature-control flag not explicitly stated for claude-fable-5."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/models/overview — reliable knowledge / training cutoff not published in the Fable 5 / Mythos 5 launch table."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15"
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for claude-fable-5; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$10.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15"
      },
      "price_output": {
        "value": "$50.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-read rate for claude-fable-5 not captured numerically in fetched sources."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-write rate for claude-fable-5 not captured numerically in fetched sources."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input × advertised context once both are confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15",
        "note": "Vision (image-in) is the only non-text input listed; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.anthropic.com/news/claude-fable-5-mythos-5 — benchmark numbers not captured from a primary leaderboard in this pass."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.anthropic.com/news/claude-fable-5-mythos-5 — SWE-bench Verified score not captured from a primary source in this pass."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for claude-fable-5 to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — score for claude-fable-5 to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for claude-fable-5 to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for claude-fable-5 to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for claude-fable-5 to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for claude-fable-5 to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for claude-fable-5 to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for claude-fable-5 in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/τ²-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
      "source_type": "provider_card",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5",
        "last_verified": "2026-06-15"
      }
    },
    {
      "id": "claude-opus-4-8",
      "name": "Claude Opus 4.8",
      "vendor": "Anthropic",
      "model_id": "claude-opus-4-8",
      "context_window": "1M",
      "max_output": "128K",
      "input_per_mtok": "$5.00",
      "output_per_mtok": "$25.00",
      "strengths": "Most capable Opus-tier model: state-of-the-art long-horizon agentic execution, knowledge work and memory. 1M context at standard pricing.",
      "provider": "Anthropic",
      "family": "Claude Opus",
      "release_date": {
        "value": "2026-05-28",
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-8.html",
        "last_verified": "2026-06-15",
        "note": "AWS Bedrock model card: 'Model launch date: May 28, 2026'."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Opus 4.8."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Opus 4.8."
      },
      "tool_call": {
        "value": true,
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-8.html",
        "last_verified": "2026-06-15",
        "note": "Bedrock card documents computer-use tool (computer_20251124) and prompt caching over system, messages and tools."
      },
      "reasoning": {
        "value": "always",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Overview table: Adaptive thinking Yes; Extended thinking No. Bedrock card lists Reasoning: Supported. Effort defaults to high."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/models/overview — schema-guaranteed structured-output flag not explicitly published for claude-opus-4-8 (no models.dev entry yet)."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-8.html",
        "last_verified": "2026-06-15",
        "note": "Bedrock input modalities: Text yes, Image yes; Audio/Speech/Video no."
      },
      "temperature": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/models/overview — temperature-control flag not explicitly stated for claude-opus-4-8."
      },
      "knowledge_cutoff": {
        "value": "2026-01",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Reliable knowledge cutoff Jan 2026; training data cutoff Jan 2026 (Bedrock card: 'Knowledge cutoff: January 2026')."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "1M tokens on Claude API; 200k on Microsoft Foundry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for claude-opus-4-8; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$5.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_output": {
        "value": "$25.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-read rate for claude-opus-4-8 not captured numerically in fetched sources."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-write rate for claude-opus-4-8 not captured numerically in fetched sources."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build: price_input × advertised context once cache context is confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-8.html",
        "last_verified": "2026-06-15"
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.anthropic.com/news — benchmark number not captured from a primary leaderboard in this pass."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for claude-opus-4-8 to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for claude-opus-4-8 to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — score for claude-opus-4-8 to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for claude-opus-4-8 to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for claude-opus-4-8 to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for claude-opus-4-8 to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for claude-opus-4-8 to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for claude-opus-4-8 (overview lists Comparative latency: Moderate)."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for claude-opus-4-8 in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/τ²-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://platform.claude.com/docs/en/about-claude/models/overview",
      "source_type": "provider_card",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "128k synchronous; up to 300k via output-300k-2026-03-24 beta on the Batches API."
      }
    },
    {
      "id": "claude-sonnet-4-6",
      "name": "Claude Sonnet 4.6",
      "vendor": "Anthropic",
      "model_id": "claude-sonnet-4-6",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$3.00",
      "output_per_mtok": "$15.00",
      "strengths": "Best balance of speed and intelligence for high-volume production agents. Adaptive thinking; 1M context.",
      "provider": "Anthropic",
      "family": "Claude Sonnet",
      "release_date": {
        "value": "2026-02-17",
        "source": "https://models.dev/models/anthropic/claude-sonnet-4-6/",
        "last_verified": "2026-06-15",
        "note": "models.dev last_updated/release for claude-sonnet-4-6."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Sonnet 4.6."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Sonnet 4.6."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-sonnet-4-6: tool_call true."
      },
      "reasoning": {
        "value": "always",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Overview: Extended thinking Yes and Adaptive thinking Yes; models.dev reasoning true."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-sonnet-4-6: structured_output true."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev input modalities: text, image."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-sonnet-4-6: temperature true."
      },
      "knowledge_cutoff": {
        "value": "2025-08-31",
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge 2025-08-31; Anthropic overview lists reliable knowledge cutoff Aug 2025, training cutoff Jan 2026."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for claude-sonnet-4-6; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$3.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_output": {
        "value": "$15.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-read rate for claude-sonnet-4-6 not captured numerically in fetched sources."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-write rate for claude-sonnet-4-6 not captured numerically in fetched sources."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build: price_input × advertised context (note tiered pricing above 200k tokens per models.dev)."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15"
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for claude-sonnet-4-6 to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for claude-sonnet-4-6 to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for claude-sonnet-4-6 to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — score for claude-sonnet-4-6 to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for claude-sonnet-4-6 to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for claude-sonnet-4-6 to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for claude-sonnet-4-6 to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for claude-sonnet-4-6 (overview Comparative latency: Fast)."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for claude-sonnet-4-6 to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for claude-sonnet-4-6 in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool, R_ctx, R_cost and R_latency inputs not yet sourced (structured_output IS sourced → R_struct=1.0). Compute per /models/agent-readiness-score once remaining inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://platform.claude.com/docs/en/about-claude/models/overview",
      "source_type": "provider_card",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Overview table: 64k tokens (Messages API). models.dev lists 128k; the Anthropic overview is treated as authoritative and the live record's 64K is preserved."
      }
    },
    {
      "id": "claude-haiku-4-5",
      "name": "Claude Haiku 4.5",
      "vendor": "Anthropic",
      "model_id": "claude-haiku-4-5",
      "context_window": "200K",
      "max_output": "64K",
      "input_per_mtok": "$1.00",
      "output_per_mtok": "$5.00",
      "strengths": "Fastest and most cost-effective Claude model — ideal for subagents, classification and latency-critical steps.",
      "provider": "Anthropic",
      "family": "Claude Haiku",
      "release_date": {
        "value": "2025-10-15",
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-haiku-4-5 release/last_updated 2025-10-15."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Haiku 4.5."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Anthropic does not disclose parameter counts for Claude Haiku 4.5."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-haiku-4-5: tool_call true."
      },
      "reasoning": {
        "value": "optional",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Overview: Extended thinking Yes, Adaptive thinking No; models.dev reasoning true. Reasoning is available but not always-on."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/api.json — structured_output flag not present on the anthropic/claude-haiku-4-5 entry in the fetched data; confirm against provider card at build."
      },
      "attachment": {
        "value": [
          "image",
          "pdf"
        ],
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev input modalities: text, image, pdf."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev anthropic/claude-haiku-4-5: temperature true."
      },
      "knowledge_cutoff": {
        "value": "2025-02-01",
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge 2025-02-01; Anthropic overview lists reliable knowledge cutoff Feb 2025, training cutoff Jul 2025."
      },
      "context_advertised": {
        "value": "200K",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for claude-haiku-4-5; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_output": {
        "value": "$5.00",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-read rate for claude-haiku-4-5 not captured numerically in fetched sources."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://platform.claude.com/docs/en/about-claude/pricing — cache-write rate for claude-haiku-4-5 not captured numerically in fetched sources."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build: price_input × advertised context (200K)."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "pdf-in",
          "text-out"
        ],
        "source": "https://models.dev/api.json",
        "last_verified": "2026-06-15"
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for claude-haiku-4-5 to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for claude-haiku-4-5 to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for claude-haiku-4-5 to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — score for claude-haiku-4-5 to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for claude-haiku-4-5 to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for claude-haiku-4-5 to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for claude-haiku-4-5 to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for claude-haiku-4-5 (overview Comparative latency: Fastest)."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for claude-haiku-4-5 to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for claude-haiku-4-5 in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool, R_struct, R_ctx, R_cost and R_latency inputs not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://platform.claude.com/docs/en/about-claude/models/overview",
      "source_type": "provider_card",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://platform.claude.com/docs/en/about-claude/models/overview",
        "last_verified": "2026-06-15",
        "note": "Overview table: 64k tokens. models.dev lists 62k; the Anthropic overview is authoritative and the live record's 64K is preserved."
      }
    },
    {
      "id": "gpt-frontier",
      "name": "GPT (frontier tier)",
      "vendor": "OpenAI",
      "model_id": "see provider",
      "context_window": "see provider",
      "max_output": "see provider",
      "input_per_mtok": "see provider",
      "output_per_mtok": "see provider",
      "strengths": "OpenAI's flagship reasoning family. Pricing and exact context vary by released variant — check OpenAI's pricing page for current numbers.",
      "provider": "OpenAI",
      "family": "GPT",
      "release_date": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs — this row pins no specific GPT variant; record the release date of the pinned variant at build."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://openai.com — OpenAI's flagship GPT API models are not open-weight, but open-weight gpt-oss variants exist; confirm per pinned variant at build."
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15",
        "note": "OpenAI API GPT models are served under OpenAI's commercial terms (proprietary); a pinned open-weight gpt-oss variant would differ."
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "OpenAI does not disclose parameter counts for its flagship GPT API models."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "OpenAI does not disclose parameter counts for its flagship GPT API models."
      },
      "tool_call": {
        "value": true,
        "source": "https://developers.openai.com/api/docs/guides/function-calling",
        "last_verified": "2026-06-15",
        "note": "Family-level: function calling (tool calling) is documented as supported across OpenAI's flagship GPT API models via the tools parameter / Responses API. Satisfies the selection gate."
      },
      "reasoning": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs — reasoning mode varies by variant (reasoning-series vs standard GPT); confirm per pinned variant at build."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs/guides/structured-outputs — Structured Outputs are documented for OpenAI models; confirm coverage for the pinned variant at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs — input modalities vary by variant; confirm per pinned variant at build."
      },
      "temperature": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs — temperature support varies (reasoning models restrict it); confirm per pinned variant at build."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs/models — knowledge cutoff varies by variant; confirm per pinned variant at build."
      },
      "context_advertised": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs/models — context window varies by variant; deferred to provider per the live record's discipline."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No measured effective-context value sourced for a pinned OpenAI variant."
      },
      "price_input": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://openai.com/api/pricing/ — input price varies by variant; deferred to provider (we will not print a price we cannot vouch for)."
      },
      "price_output": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://openai.com/api/pricing/ — output price varies by variant; deferred to provider."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://openai.com/api/pricing/ — cached-input price varies by variant; confirm at build."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://openai.com/api/pricing/ — confirm cache-write semantics/price per variant at build."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derivable only once a pinned variant's price + context are confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derivable only once a pinned variant's cache-aware pricing is confirmed."
      },
      "modalities": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://developers.openai.com/api/docs/models — modality arrays vary by variant; confirm per pinned variant at build."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned GPT variant at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — per pinned GPT variant at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — per pinned GPT variant at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — per pinned GPT variant at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — per pinned GPT variant at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned GPT variant at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — per pinned GPT variant at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned GPT variant at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned GPT variant at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Per pinned GPT variant at build from a primary source."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: row pins no specific GPT variant, so no input beyond tool_call is sourced. Pin a variant, then compute per /models/agent-readiness-score."
      },
      "score_confidence": "partial",
      "source_url": "https://developers.openai.com/api/docs/guides/function-calling",
      "source_type": "provider_card",
      "last_verified": "2026-06-15"
    },
    {
      "id": "gemini-frontier",
      "name": "Gemini (frontier tier)",
      "vendor": "Google",
      "model_id": "see provider",
      "context_window": "1M+ (varies)",
      "max_output": "see provider",
      "input_per_mtok": "see provider",
      "output_per_mtok": "see provider",
      "strengths": "Long-context multimodal family; some variants advertise multi-million-token windows. Confirm pricing on Google's pricing page.",
      "provider": "Google",
      "family": "Gemini",
      "release_date": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/models — this row pins no specific Gemini variant; record the release date of the pinned variant at build."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://ai.google.dev/gemini-api/docs/models",
        "last_verified": "2026-06-15",
        "note": "The Gemini API family is proprietary/closed-weight (distinct from Google's open Gemma family)."
      },
      "license": {
        "value": "proprietary",
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15",
        "note": "Gemini API models are served under Google's API terms (proprietary)."
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Google does not disclose parameter counts for Gemini API models."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Google does not disclose parameter counts for Gemini API models."
      },
      "tool_call": {
        "value": true,
        "source": "https://ai.google.dev/gemini-api/docs/function-calling",
        "last_verified": "2026-06-15",
        "note": "Family-level: function calling is documented as supported across Gemini API models. Satisfies the selection gate."
      },
      "reasoning": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs — thinking/reasoning support varies by variant; confirm per pinned variant at build."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/structured-output — structured output is documented for Gemini; confirm coverage for the pinned variant at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs — Gemini is multimodal but exact accepted modalities vary by variant; confirm per pinned variant at build."
      },
      "temperature": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs — confirm temperature exposure per pinned variant at build."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/models — knowledge cutoff varies by variant; confirm per pinned variant at build."
      },
      "context_advertised": {
        "value": "1M+ (varies)",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/models — preserves the live record's deferred value; some variants advertise multi-million-token windows. Confirm exact number per pinned variant at build."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No measured effective-context value sourced for a pinned Gemini variant."
      },
      "price_input": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/pricing — input price varies by variant; deferred to provider (we will not print a price we cannot vouch for)."
      },
      "price_output": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/pricing — output price varies by variant; deferred to provider."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/pricing — context-cache price varies by variant; confirm at build."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/pricing — confirm cache storage/write semantics per variant at build."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derivable only once a pinned variant's price + context are confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derivable only once a pinned variant's cache-aware pricing is confirmed."
      },
      "modalities": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://ai.google.dev/gemini-api/docs/models — multimodal arrays vary by variant; confirm per pinned variant at build."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Gemini variant at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — per pinned Gemini variant at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — per pinned Gemini variant at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — per pinned Gemini variant at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — per pinned Gemini variant at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Gemini variant at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — per pinned Gemini variant at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Gemini variant at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Gemini variant at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Per pinned Gemini variant at build from a primary source."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: row pins no specific Gemini variant, so no input beyond tool_call is sourced. Pin a variant, then compute per /models/agent-readiness-score."
      },
      "score_confidence": "partial",
      "source_url": "https://ai.google.dev/gemini-api/docs/function-calling",
      "source_type": "provider_card",
      "last_verified": "2026-06-15"
    },
    {
      "id": "llama-open",
      "name": "Llama (open weights)",
      "vendor": "Meta",
      "model_id": "see provider",
      "context_window": "varies",
      "max_output": "varies",
      "input_per_mtok": "self-host or per-host",
      "output_per_mtok": "self-host or per-host",
      "strengths": "Open-weights family you can run yourself; effective price depends on your inference host, not a list price.",
      "provider": "Meta",
      "family": "Llama",
      "release_date": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/ — this row pins no specific Llama version; record the release date of the pinned version at build."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://www.llama.com/llama4/license/",
        "last_verified": "2026-06-15",
        "note": "Llama 4 weights are openly downloadable under the Llama 4 Community License Agreement."
      },
      "license": {
        "value": "Llama 4 Community License Agreement",
        "source": "https://www.llama.com/llama4/license/",
        "last_verified": "2026-06-15",
        "note": "Source-available community license (not OSI-approved): royalty-free use/redistribution with a 700M-MAU commercial threshold requiring a separate Meta license. Applies to Llama 4; a different pinned version may carry a different license."
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/ — total parameter count varies by pinned Llama variant; record at build."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/ — active (MoE) parameter count varies by pinned Llama variant; record at build."
      },
      "tool_call": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/docs/ — Llama 4 supports tool calling, but this row pins no specific version and tool-call behavior is host/template-dependent; confirm per pinned variant + host at build. (Selection-gate attribute — must resolve true for inclusion.)"
      },
      "reasoning": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/docs/ — reasoning support varies by pinned Llama variant; confirm at build."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/docs/ — structured-output support is host/framework-dependent for open weights; confirm per pinned variant + host at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/docs/ — multimodal input varies by pinned Llama variant; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://www.llama.com/docs/",
        "last_verified": "2026-06-15",
        "note": "Temperature is a standard inference parameter for self-hosted open-weight Llama models; exposed by the host runtime."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/ — knowledge cutoff varies by pinned Llama variant; confirm at build."
      },
      "context_advertised": {
        "value": "varies",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/docs/ — context window varies by pinned Llama variant; preserves the live record's deferred value."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No measured effective-context value sourced for a pinned Llama variant."
      },
      "price_input": {
        "value": "self-host or per-host",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Open weights: effective input price depends on the inference host, not a Meta list price (preserves the live record's discipline)."
      },
      "price_output": {
        "value": "self-host or per-host",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Open weights: effective output price depends on the inference host, not a Meta list price."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Prompt-cache pricing is host-defined for self-hosted Llama; confirm per host at build."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Prompt-cache pricing is host-defined for self-hosted Llama; confirm per host at build."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Not a list price for open weights; depends on host + chosen GPU economics."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Not a list price for open weights; depends on host + chosen GPU economics."
      },
      "modalities": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.llama.com/ — modality arrays vary by pinned Llama variant; confirm at build."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Llama variant at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — per pinned Llama variant at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — per pinned Llama variant at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary τ²-Bench leaderboard — per pinned Llama variant at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — per pinned Llama variant at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — per pinned Llama variant at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — per pinned Llama variant at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Throughput is host/hardware-dependent for self-hosted Llama; confirm per host at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "TTFT is host/hardware-dependent for self-hosted Llama; confirm per host at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Per pinned Llama variant at build from a primary source."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: row pins no specific Llama version and most inputs are host-dependent. Pin a variant + host, then compute per /models/agent-readiness-score."
      },
      "score_confidence": "partial",
      "source_url": "https://www.llama.com/llama4/license/",
      "source_type": "provider_card",
      "last_verified": "2026-06-15"
    },
    {
      "id": "gpt-5",
      "name": "GPT-5",
      "vendor": "OpenAI",
      "model_id": "gpt-5",
      "context_window": "400K",
      "max_output": "128K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$10.00",
      "strengths": "OpenAI's flagship reasoning model: 400K context, native tool calling and schema-guaranteed structured output. A frontier agentic workhorse.",
      "provider": "OpenAI",
      "family": "GPT-5",
      "release_date": {
        "value": "2025-08-07",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-09",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "400K",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$10.00",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/gpt-5/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/openai/gpt-5/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gpt-5-1",
      "name": "GPT-5.1",
      "vendor": "OpenAI",
      "model_id": "gpt-5.1",
      "context_window": "400K",
      "max_output": "128K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$10.00",
      "strengths": "Refreshed GPT-5 flagship (Nov 2025): same 400K context and tool calling, tuned for agentic workflows.",
      "provider": "OpenAI",
      "family": "GPT-5",
      "release_date": {
        "value": "2025-11-13",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-09",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "400K",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$10.00",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1/ — modality array not enumerated on the fetched entry; confirm at build."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/gpt-5.1/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/openai/gpt-5.1/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "models.dev marks structured_output absent on the gpt-5.1 entry (gpt-5 and gpt-5.1-codex mark it true); kept as placeholder rather than guessed."
    },
    {
      "id": "gpt-5-mini",
      "name": "GPT-5 Mini",
      "vendor": "OpenAI",
      "model_id": "gpt-5-mini",
      "context_window": "400K",
      "max_output": "128K",
      "input_per_mtok": "$0.25",
      "output_per_mtok": "$2.00",
      "strengths": "Cost-efficient GPT-5 tier for high-volume agents and subagents: 400K context, tool calling and structured output at a fraction of flagship price.",
      "provider": "OpenAI",
      "family": "GPT-5",
      "release_date": {
        "value": "2025-08-07",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-mini/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-mini/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-05",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "400K",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.25",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$2.00",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-mini/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-mini/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/gpt-5-mini/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/openai/gpt-5-mini/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gpt-5-codex",
      "name": "GPT-5 Codex",
      "vendor": "OpenAI",
      "model_id": "gpt-5-codex",
      "context_window": "400K",
      "max_output": "128K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$10.00",
      "strengths": "Coding-agent specialization of GPT-5: 400K context, tool calling and structured output, tuned for software-engineering loops.",
      "provider": "OpenAI",
      "family": "GPT-5",
      "release_date": {
        "value": "2025-09-15",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-codex/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-codex/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-codex/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-09",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "400K",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$10.00",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-codex/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5-codex/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/gpt-5-codex/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/openai/gpt-5-codex/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gpt-5-1-codex",
      "name": "GPT-5.1 Codex",
      "vendor": "OpenAI",
      "model_id": "gpt-5.1-codex",
      "context_window": "400K",
      "max_output": "128K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$10.00",
      "strengths": "Coding-agent specialization of GPT-5.1: 400K context, tool calling and structured output for SWE agents.",
      "provider": "OpenAI",
      "family": "GPT-5",
      "release_date": {
        "value": "2025-11-13",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1-codex/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1-codex/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1-codex/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-09",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "400K",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$10.00",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1-codex/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/gpt-5.1-codex/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/gpt-5.1-codex/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/openai/gpt-5.1-codex/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "openai-o3",
      "name": "OpenAI o3",
      "vendor": "OpenAI",
      "model_id": "o3",
      "context_window": "200K",
      "max_output": "100K",
      "input_per_mtok": "$2.00",
      "output_per_mtok": "$8.00",
      "strengths": "Dedicated reasoning model with tool calling and structured output: deep multi-step problem solving for analytical agents.",
      "provider": "OpenAI",
      "family": "OpenAI o-series",
      "release_date": {
        "value": "2025-04-16",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://openai.com/policies/services-agreement/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/o3/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/o3/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image"
        ],
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": false,
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-05",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "200K",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$2.00",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$8.00",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/o3/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/openai/o3/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "text-out"
        ],
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/openai/o3/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "100K",
        "source": "https://models.dev/models/openai/o3/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gemini-3-pro",
      "name": "Gemini 3 Pro",
      "vendor": "Google",
      "model_id": "gemini-3-pro-preview",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$2.00",
      "output_per_mtok": "$12.00",
      "strengths": "Google's frontier long-context multimodal model: ~1M-token window, thinking, tool calling and structured output.",
      "provider": "Google",
      "family": "Gemini 3",
      "release_date": {
        "value": "2025-11-18",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-pro-preview/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-pro-preview/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image",
          "audio",
          "video",
          "pdf"
        ],
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-01",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$2.00",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$12.00",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-pro-preview/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-pro-preview/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "audio-in",
          "video-in",
          "pdf-in",
          "text-out"
        ],
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/google/gemini-3-pro-preview/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/google/gemini-3-pro-preview/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "max_output 65,536 tokens on models.dev rendered as 64K to match the dataset's K convention."
    },
    {
      "id": "gemini-3-flash",
      "name": "Gemini 3 Flash",
      "vendor": "Google",
      "model_id": "gemini-3-flash-preview",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$0.50",
      "output_per_mtok": "$3.00",
      "strengths": "Fast, cheap Gemini 3 tier with ~1M context, thinking, tool calling and structured output: built for high-throughput multimodal agents.",
      "provider": "Google",
      "family": "Gemini 3",
      "release_date": {
        "value": "2025-12-17",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-flash-preview/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-flash-preview/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image",
          "audio",
          "video",
          "pdf"
        ],
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-01",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.50",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$3.00",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-flash-preview/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-3-flash-preview/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "audio-in",
          "video-in",
          "pdf-in",
          "text-out"
        ],
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/google/gemini-3-flash-preview/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/google/gemini-3-flash-preview/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gemini-2-5-pro",
      "name": "Gemini 2.5 Pro",
      "vendor": "Google",
      "model_id": "gemini-2.5-pro",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$10.00",
      "strengths": "Proven long-context multimodal workhorse: ~1M-token window, thinking, tool calling and structured output.",
      "provider": "Google",
      "family": "Gemini 2.5",
      "release_date": {
        "value": "2025-06-17",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-pro/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-pro/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image",
          "audio",
          "video",
          "pdf"
        ],
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-01",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$10.00",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-pro/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-pro/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "audio-in",
          "video-in",
          "pdf-in",
          "text-out"
        ],
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/google/gemini-2.5-pro/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/google/gemini-2.5-pro/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "gemini-2-5-flash",
      "name": "Gemini 2.5 Flash",
      "vendor": "Google",
      "model_id": "gemini-2.5-flash",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$0.30",
      "output_per_mtok": "$2.50",
      "strengths": "High-volume multimodal agent tier: ~1M context, thinking, tool calling and structured output at low cost.",
      "provider": "Google",
      "family": "Gemini 2.5",
      "release_date": {
        "value": "2025-06-17",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://ai.google.dev/gemini-api/terms",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-flash/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-flash/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": [
          "image",
          "audio",
          "video",
          "pdf"
        ],
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities for this model."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-01",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.30",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$2.50",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-flash/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/google/gemini-2.5-flash/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "image-in",
          "audio-in",
          "video-in",
          "pdf-in",
          "text-out"
        ],
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/google/gemini-2.5-flash/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/google/gemini-2.5-flash/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "grok-4-3",
      "name": "Grok 4.3",
      "vendor": "xAI",
      "model_id": "grok-4.3",
      "context_window": "1M",
      "max_output": "30K",
      "input_per_mtok": "$1.25",
      "output_per_mtok": "$2.50",
      "strengths": "xAI's current flagship: 1M-token context, reasoning and tool calling, tuned for agentic chat and coding.",
      "provider": "xAI",
      "family": "Grok 4",
      "release_date": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — release date to confirm at build."
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://docs.x.ai/docs/models",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://docs.x.ai/docs/models",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — knowledge cutoff not published on the fetched entry; confirm at build."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.25",
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$2.50",
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/xai/grok-4.3/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/xai/grok-4.3/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "30K",
        "source": "https://models.dev/models/xai/grok-4.3/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "Required context_window (1M) and price ($1.25/$2.50) cross-confirmed by xAI's own docs (https://docs.x.ai/docs/models) AND models.dev. release_date/knowledge_cutoff left as placeholders (sources conflicted)."
    },
    {
      "id": "deepseek-chat",
      "name": "DeepSeek-V4-Flash (deepseek-chat)",
      "vendor": "DeepSeek",
      "model_id": "deepseek-chat",
      "context_window": "1M",
      "max_output": "384K",
      "input_per_mtok": "$0.14",
      "output_per_mtok": "$0.28",
      "strengths": "Non-thinking mode of DeepSeek-V4-Flash: 1M context, very low price, tool calling. The deepseek-chat API alias.",
      "provider": "DeepSeek",
      "family": "DeepSeek V4",
      "release_date": {
        "value": "2025-12-01",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://api-docs.deepseek.com/quick_start/pricing",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (MIT-class; see model card)",
        "source": "https://api-docs.deepseek.com/quick_start/pricing",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-chat/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-chat/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "none",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-chat/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-09",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.14",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$0.28",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-chat/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-chat/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/deepseek/deepseek-chat/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "384K",
        "source": "https://models.dev/models/deepseek/deepseek-chat/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "Context (1M), max output (384K) and price ($0.14/$0.28) cross-confirmed by DeepSeek's own pricing docs (api-docs.deepseek.com) AND models.dev. deepseek-chat = non-thinking mode of DeepSeek-V4-Flash per provider docs."
    },
    {
      "id": "deepseek-reasoner",
      "name": "DeepSeek-V4-Flash (deepseek-reasoner)",
      "vendor": "DeepSeek",
      "model_id": "deepseek-reasoner",
      "context_window": "1M",
      "max_output": "384K",
      "input_per_mtok": "$0.14",
      "output_per_mtok": "$0.28",
      "strengths": "Thinking mode of DeepSeek-V4-Flash: 1M context, chain-of-thought reasoning and tool calling at low cost.",
      "provider": "DeepSeek",
      "family": "DeepSeek V4",
      "release_date": {
        "value": "2025-12-01",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://api-docs.deepseek.com/quick_start/pricing",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (MIT-class; see model card)",
        "source": "https://api-docs.deepseek.com/quick_start/pricing",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-09",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.14",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$0.28",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/deepseek/deepseek-reasoner/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/deepseek/deepseek-reasoner/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "384K",
        "source": "https://models.dev/models/deepseek/deepseek-reasoner/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "Context (1M), max output (384K) and price ($0.14/$0.28) cross-confirmed by DeepSeek's own pricing docs AND models.dev. deepseek-reasoner = thinking mode of DeepSeek-V4-Flash. structured_output not marked on the canonical entry -> placeholder."
    },
    {
      "id": "qwen3-max",
      "name": "Qwen3 Max",
      "vendor": "Alibaba",
      "model_id": "qwen3-max",
      "context_window": "262K",
      "max_output": "64K",
      "input_per_mtok": "$1.20",
      "output_per_mtok": "$6.00",
      "strengths": "Alibaba's flagship Qwen3 tier: 262K context, tool calling and structured output for general agentic tasks.",
      "provider": "Alibaba",
      "family": "Qwen3",
      "release_date": {
        "value": "2025-09-23",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-max/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-max/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "none",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-max/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-04",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "262K",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.20",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$6.00",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-max/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-max/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/alibaba/qwen3-max/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/alibaba/qwen3-max/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context_advertised 262,144 / max_output 65,536 rendered as 262K / 64K to match dataset K-convention."
    },
    {
      "id": "qwen3-235b-a22b",
      "name": "Qwen3 235B-A22B",
      "vendor": "Alibaba",
      "model_id": "qwen3-235b-a22b",
      "context_window": "131K",
      "max_output": "16K",
      "input_per_mtok": "$0.10",
      "output_per_mtok": "$0.60",
      "strengths": "Open-weights Qwen3 MoE (235B total / 22B active): 131K context, reasoning and tool calling at very low cost.",
      "provider": "Alibaba",
      "family": "Qwen3",
      "release_date": {
        "value": "2025-04",
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (Apache-2.0; see model card)",
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Model name encodes 235B total parameters; confirm exact figure against the Qwen3 model card at build."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Model name (A22B) encodes ~22B active MoE parameters; confirm exact figure against the Qwen3 model card at build."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/ — knowledge cutoff not published on the fetched entry; confirm at build."
      },
      "context_advertised": {
        "value": "131K",
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "note": "models.dev lists multiple resale prices; the canonical provider price varies — see provider (do not vouch for a single number)"
      },
      "price_output": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "note": "models.dev lists multiple resale prices; the canonical provider price varies — see provider (do not vouch for a single number)"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-235b-a22b/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "16K",
        "source": "https://models.dev/models/alibaba/qwen3-235b-a22b/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "structured_output marked absent on the canonical entry -> placeholder. params hinted from the model name (235B/A22B) but left as placeholders pending the model card."
    },
    {
      "id": "qwen3-coder-plus",
      "name": "Qwen3 Coder Plus",
      "vendor": "Alibaba",
      "model_id": "qwen3-coder-plus",
      "context_window": "1M",
      "max_output": "64K",
      "input_per_mtok": "$1.00",
      "output_per_mtok": "$5.00",
      "strengths": "Coding-agent Qwen3 tier: ~1M context, tool calling and structured output for software-engineering loops.",
      "provider": "Alibaba",
      "family": "Qwen3",
      "release_date": {
        "value": "2025-07-23",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-coder-plus/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-coder-plus/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "none",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-coder-plus/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-04",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "1M",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.00",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$5.00",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-coder-plus/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/alibaba/qwen3-coder-plus/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/alibaba/qwen3-coder-plus/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "64K",
        "source": "https://models.dev/models/alibaba/qwen3-coder-plus/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context_advertised 1,048,576 / max_output 65,536 rendered as 1M / 64K."
    },
    {
      "id": "mistral-large",
      "name": "Mistral Large",
      "vendor": "Mistral",
      "model_id": "mistral-large-latest",
      "context_window": "262K",
      "max_output": "262K",
      "input_per_mtok": "$0.50",
      "output_per_mtok": "$1.50",
      "strengths": "Mistral's flagship: 262K context, tool calling and structured output for general European-sovereign agent stacks.",
      "provider": "Mistral",
      "family": "Mistral Large",
      "release_date": {
        "value": "2024-11-01",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-large-latest/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-large-latest/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "none",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-large-latest/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2024-11",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "262K",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.50",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$1.50",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-large-latest/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-large-latest/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/mistral/mistral-large-latest/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "262K",
        "source": "https://models.dev/models/mistral/mistral-large-latest/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context_advertised and max_output both 262,144 per models.dev rendered as 262K."
    },
    {
      "id": "mistral-medium",
      "name": "Mistral Medium",
      "vendor": "Mistral",
      "model_id": "mistral-medium-latest",
      "context_window": "262K",
      "max_output": "262K",
      "input_per_mtok": "$0.40",
      "output_per_mtok": "$2.00",
      "strengths": "Mid-tier Mistral: 262K context, tool calling and structured output, balanced cost for production agents.",
      "provider": "Mistral",
      "family": "Mistral Medium",
      "release_date": {
        "value": "2025-08-12",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-medium-latest/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-medium-latest/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "none",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-medium-latest/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-05",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "262K",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.40",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$2.00",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-medium-latest/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/mistral-medium-latest/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/mistral/mistral-medium-latest/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "262K",
        "source": "https://models.dev/models/mistral/mistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      }
    },
    {
      "id": "magistral-medium",
      "name": "Magistral Medium",
      "vendor": "Mistral",
      "model_id": "magistral-medium-latest",
      "context_window": "128K",
      "max_output": "16K",
      "input_per_mtok": "$2.00",
      "output_per_mtok": "$5.00",
      "strengths": "Mistral's reasoning model: 128K context, chain-of-thought reasoning and tool calling.",
      "provider": "Mistral",
      "family": "Magistral",
      "release_date": {
        "value": "2025-03-17",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": false,
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "proprietary",
        "source": "https://mistral.ai/terms/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-06",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "128K",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$2.00",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$5.00",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/mistral/magistral-medium-latest/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/mistral/magistral-medium-latest/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "16K",
        "source": "https://models.dev/models/mistral/magistral-medium-latest/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "structured_output shown as '-' on the canonical entry -> placeholder."
    },
    {
      "id": "glm-5",
      "name": "GLM-5",
      "vendor": "Zhipu AI",
      "model_id": "glm-5",
      "context_window": "200K",
      "max_output": "128K",
      "input_per_mtok": "$1.00",
      "output_per_mtok": "$3.20",
      "strengths": "Zhipu's open-weights flagship: ~200K context, reasoning and tool calling, agentic-oriented.",
      "provider": "Zhipu AI",
      "family": "GLM-5",
      "release_date": {
        "value": "2026-02-11",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (see Hugging Face zai-org/GLM-5)",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — knowledge cutoff not published on the fetched entry; confirm at build."
      },
      "context_advertised": {
        "value": "200K",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$1.00",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$3.20",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-5/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/zhipuai/glm-5/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/zhipuai/glm-5/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context 204,800 / max output 131,072 rendered as 200K / 128K. structured_output shown as '-' on canonical Zhipu provider -> placeholder."
    },
    {
      "id": "glm-4-7",
      "name": "GLM-4.7",
      "vendor": "Zhipu AI",
      "model_id": "glm-4.7",
      "context_window": "200K",
      "max_output": "128K",
      "input_per_mtok": "$0.60",
      "output_per_mtok": "$2.20",
      "strengths": "Open-weights GLM-4.7: ~200K context, reasoning, tool calling and structured output at low cost.",
      "provider": "Zhipu AI",
      "family": "GLM-4",
      "release_date": {
        "value": "2025-12-22",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (see Hugging Face)",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.7/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.7/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.7/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": "2025-04",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "models.dev knowledge cutoff for this model."
      },
      "context_advertised": {
        "value": "200K",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "$0.60",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider input price per the models.dev entry; resale prices vary."
      },
      "price_output": {
        "value": "$2.20",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Canonical-provider output price per the models.dev entry; resale prices vary."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.7/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.7/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/zhipuai/glm-4.7/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/zhipuai/glm-4.7/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context 204,800 / max output 131,072 rendered as 200K / 128K."
    },
    {
      "id": "glm-4-6",
      "name": "GLM-4.6",
      "vendor": "Zhipu AI",
      "model_id": "glm-4.6",
      "context_window": "200K",
      "max_output": "128K",
      "input_per_mtok": "$0.43",
      "output_per_mtok": "$1.74",
      "strengths": "Open-weights GLM-4.6: ~200K context, reasoning and tool calling, a low-cost agentic workhorse.",
      "provider": "Zhipu AI",
      "family": "GLM-4",
      "release_date": {
        "value": "2025-09-30",
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (see Hugging Face)",
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — structured_output not marked on the canonical provider entry; confirm at build."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — knowledge cutoff not published on the fetched entry; confirm at build."
      },
      "context_advertised": {
        "value": "200K",
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/",
        "note": "models.dev lists multiple resale prices; the canonical provider price varies — see provider (do not vouch for a single number)"
      },
      "price_output": {
        "value": "see provider",
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/",
        "note": "models.dev lists multiple resale prices; the canonical provider price varies — see provider (do not vouch for a single number)"
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/zhipuai/glm-4.6/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/zhipuai/glm-4.6/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "128K",
        "source": "https://models.dev/models/zhipuai/glm-4.6/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context 204,800 / max output 131,072 rendered as 200K / 128K. structured_output shown as '-' on canonical entry -> placeholder."
    },
    {
      "id": "kimi-k2",
      "name": "Kimi K2",
      "vendor": "Moonshot AI",
      "model_id": "kimi-k2",
      "context_window": "262K",
      "max_output": "262K",
      "input_per_mtok": "see provider",
      "output_per_mtok": "see provider",
      "strengths": "Moonshot's open-weights agentic model: 262K context, reasoning, tool calling and structured output.",
      "provider": "Moonshot AI",
      "family": "Kimi K2",
      "release_date": {
        "value": "2026-04-21",
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15"
      },
      "last_updated": "2026-06-15",
      "open_weights": {
        "value": true,
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15"
      },
      "license": {
        "value": "open (see Hugging Face)",
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15"
      },
      "params_total": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — total parameter count not published for this model."
      },
      "params_active": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — active (MoE) parameter count not published for this model."
      },
      "tool_call": {
        "value": true,
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "models.dev capability flag tool_call: true. Satisfies the selection gate."
      },
      "reasoning": {
        "value": "always",
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "Derived from the models.dev reasoning flag for this model."
      },
      "structured_output": {
        "value": true,
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "models.dev structured_output flag for this model."
      },
      "attachment": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — accepted attachment/file modalities not enumerated on the fetched entry; confirm at build."
      },
      "temperature": {
        "value": true,
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "models.dev temperature flag for this model."
      },
      "knowledge_cutoff": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — knowledge cutoff not published on the fetched entry; confirm at build."
      },
      "context_advertised": {
        "value": "262K",
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "Advertised input context window per the canonical models.dev provider entry."
      },
      "context_effective": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No published measured long-context recall benchmark for this model; do not derive effective context from advertised."
      },
      "price_input": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — canonical input price not reliably resolvable (placeholder pricing on the entry); see provider."
      },
      "price_output": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — canonical output price not reliably resolvable (placeholder pricing on the entry); see provider."
      },
      "price_cache_read": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — cache-read rate not captured numerically from a primary source in this pass."
      },
      "price_cache_write": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://models.dev/models/moonshotai/kimi-k2/ — cache-write rate not captured numerically from a primary source in this pass."
      },
      "cost_per_full_window": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from price_input x advertised context once confirmed."
      },
      "cost_per_agent_task": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Derived at build from the cache-aware agent-task model once cache rates are confirmed."
      },
      "modalities": {
        "value": [
          "text-in",
          "text-out"
        ],
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "Derived from models.dev input modalities; output is text."
      },
      "gpqa_diamond": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — GPQA-Diamond for this model to confirm at build."
      },
      "swe_bench_verified": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.swebench.com/ — SWE-bench Verified score for this model to confirm at build."
      },
      "terminal_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://www.tbench.ai/ — Terminal-Bench score for this model to confirm at build."
      },
      "tau2_bench": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Primary tau2-Bench leaderboard — score for this model to confirm at build."
      },
      "bfcl_tool_use": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://gorilla.cs.berkeley.edu/leaderboard.html — BFCL tool-use score for this model to confirm at build."
      },
      "aa_index": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — Intelligence Index for this model to confirm at build."
      },
      "lmarena_elo": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://lmarena.ai/leaderboard — human-preference Elo for this model to confirm at build."
      },
      "tokens_per_sec": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured throughput for this model to confirm at build."
      },
      "ttft": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "https://artificialanalysis.ai/ — measured time-to-first-token for this model to confirm at build."
      },
      "hallucination_rate": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "No primary hallucination benchmark captured for this model in this pass."
      },
      "agent_readiness_score": {
        "value": null,
        "verify_status": "verify-against-primary-at-build",
        "source_hint": "Score withheld: R_tool (BFCL/tau2-Bench), R_ctx (effective context), R_cost (cache rates) and R_latency inputs are not yet sourced. Compute per /models/agent-readiness-score once inputs confirmed."
      },
      "score_confidence": "partial",
      "source_url": "https://models.dev/models/moonshotai/kimi-k2/",
      "source_type": "models.dev",
      "last_verified": "2026-06-15",
      "max_output_meta": {
        "value": "262K",
        "source": "https://models.dev/models/moonshotai/kimi-k2/",
        "last_verified": "2026-06-15",
        "note": "Max output tokens per the canonical models.dev provider entry."
      },
      "_provenance_note": "context_advertised and max_output both 262,144 per models.dev. Canonical entry showed $0.00/$0.00 (placeholder) -> price deferred to 'see provider' rather than asserting a free price."
    }
  ]
}