{
  "meta": {
    "api_version": "v1",
    "endpoint": "/api/v1/changes",
    "lifecycle_updated": "2026-09-19",
    "total": 76,
    "offset": 0,
    "limit": 50,
    "returned": 50,
    "provenance": "Combines append-only history snapshots with dated deprecation announcements from the sourced lifecycle record; source_layer identifies the origin."
  },
  "data": [
    {
      "date": "2026-09-17",
      "model_id": "antigravity-preview-05-2026",
      "model_name": "Antigravity Agent 05-2026",
      "provider": "Google",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-10-05",
      "replacement_official": {
        "name": "Antigravity Agent 09-2026",
        "api_id": "antigravity-preview-09-2026"
      },
      "source": "https://ai.google.dev/gemini-api/docs/changelog",
      "verified": "2026-09-19",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-09-10",
      "model_id": "deepseek-v4-flash",
      "model_name": "DeepSeek-V4-Flash and V4-Flash-Vision-Exp",
      "provider": "DeepSeek",
      "event": "deprecation_announced",
      "lifecycle_status": "retired",
      "shutdown": "2026-09-10",
      "replacement_official": {
        "name": "DeepSeek-V4.1-Flash",
        "api_id": "deepseek-flash"
      },
      "source": "https://api-docs.deepseek.com/news/news260910",
      "verified": "2026-09-11",
      "source_layer": "deprecations"
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-09-03",
      "model_id": "gpt-6-astra",
      "model_name": "GPT-6 Astra",
      "event": "new_model",
      "pricing": {
        "input_per_million": 10,
        "output_per_million": 50
      },
      "context_max_tokens": 1050000,
      "note": "OpenAI announced gpt-6-astra in its API changelog on September 3, 2026 as its most capable model, for reasoning, coding, computer use, research and document creation. The model directory lists a 1.05M context window and an April 30, 2026 knowledge cutoff, and publishes no maximum-output figure. Pricing is $10.00 input / $1.00 cached input / $50.00 output per 1M tokens, with Batch at half of each. The changelog records interface constraints that are migration blockers: no `none` reasoning-effort level, no custom temperature, top_p or logprobs, and tool calling only through the Responses API. Source: developers.openai.com/api/docs/changelog, /pricing and /models (verified 2026-09-08)."
    },
    {
      "source_layer": "history",
      "provider": "Google",
      "date": "2026-09-02",
      "model_id": "gemini-3-8-flash",
      "model_name": "Gemini 3.8 Flash",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.75,
        "output_per_million": 3.75
      },
      "context_max_tokens": 1048576,
      "note": "Google released gemini-3.8-flash on September 2, 2026, its third Flash release in six weeks. Official docs list a 1,048,576-token input limit, 65,536 max output, and tunable thinking levels. Introductory pricing is $0.75/$3.75 per 1M input/output through December 31, 2026, doubling to $1.50/$7.50 on January 1, 2027; batch and flex are $0.375/$1.875, priority $1.35/$6.75, cached input $0.075. Google published HLE-Verified 54.9% and a 47.2% pass@1 patching figure in the announcement, neither of which maps to a benchmark column benchr tracks, so no benchmark value is recorded. The 3.8 Flash Cyber variant is Fairwind Program only and is not separately priced. Google states 3.7 Flash remains fully supported. Sources: blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/ and ai.google.dev/gemini-api/docs/pricing (verified 2026-09-03)."
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-09-01",
      "model_id": "claude-fable-5-1",
      "model_name": "Claude Fable 5.1",
      "event": "new_model",
      "pricing": {
        "input_per_million": 10,
        "output_per_million": 50
      },
      "context_max_tokens": 1000000,
      "note": "Anthropic released claude-fable-5-1 on September 1, 2026. Official docs list 1M context, 128K max output, $10/$50 standard input/output per 1M, $0.25 cached input, and Batch at $5/$25. Adaptive thinking is always on. Source: anthropic.com/claude-fable-and-mythos-5-1 and platform.claude.com/docs/en/models/fable-5-1/overview (verified 2026-09-02)."
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-08-31",
      "model_id": "claude-sonnet-5",
      "model_name": "Claude Sonnet 5",
      "event": "price_change",
      "pricing": {
        "input_per_million": 2,
        "output_per_million": 10
      },
      "note": "Anthropic's pricing page now lists $2/$10 per 1M as Claude Sonnet 5's standard price. The increase to $3/$15 scheduled for September 1, 2026 was cancelled; cache-hit input stays $0.20. Verified 2026-08-31."
    },
    {
      "source_layer": "history",
      "provider": "DeepSeek",
      "date": "2026-08-28",
      "model_id": "deepseek-v4-flash",
      "model_name": "DeepSeek V4-Flash",
      "event": "price_change",
      "pricing": {
        "input_per_million": 0.44,
        "output_per_million": 1.32
      },
      "context_max_tokens": 1000000,
      "note": "Cache-miss input rose from a flat $0.14 to $0.44 peak / $0.22 off-peak per 1M tokens, output from $0.28 to $1.32 / $0.66, and cache-hit input from $0.0028 to $0.014 / $0.007. DeepSeek replaced flat API pricing with peak/off-peak billing at 16:00 UTC on August 16, 2026. Peak hours are 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday; all other hours bill at half the peak rate. The pricing figures here are the peak (list) rates. Source: api-docs.deepseek.com/quick_start/pricing (read 2026-08-28)."
    },
    {
      "source_layer": "history",
      "provider": "DeepSeek",
      "date": "2026-08-28",
      "model_id": "deepseek-v4-pro",
      "model_name": "DeepSeek V4-Pro",
      "event": "price_change",
      "pricing": {
        "input_per_million": 1.32,
        "output_per_million": 3.96
      },
      "context_max_tokens": 1000000,
      "note": "Cache-miss input rose from a flat $0.435 to $1.32 peak / $0.66 off-peak per 1M tokens, output from $0.87 to $3.96 / $1.98, and cache-hit input from $0.003625 to $0.044 / $0.022. DeepSeek replaced flat API pricing with peak/off-peak billing at 16:00 UTC on August 16, 2026. Peak hours are 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday; all other hours bill at half the peak rate. The pricing figures here are the peak (list) rates. Source: api-docs.deepseek.com/quick_start/pricing (read 2026-08-28)."
    },
    {
      "source_layer": "history",
      "provider": "Alibaba (Qwen)",
      "date": "2026-08-27",
      "model_id": "qwen-3-8-flash",
      "model_name": "Qwen3.8-Flash",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.15,
        "output_per_million": 0.47
      },
      "context_max_tokens": 262144,
      "note": "Alibaba released Qwen3.8-Flash on August 27, 2026: a multimodal mixture-of-experts model with a 125B main model plus 51B of N-gram embeddings, activating 6B parameters per token, with a native 262K context the announcement says extends to 1M. Model Studio lists $0.15 input / $0.47 output per 1M tokens in the Singapore region. Alibaba names SWE-bench Pro, CoWorkBench, Toolathlon Verified, MathVision, AndroidWorld and ERQA but publishes no scores, and names no license. Source: alibabacloud.com/blog/...603503 and alibabacloud.com/help/en/model-studio/model-pricing (verified 2026-09-08)."
    },
    {
      "source_layer": "history",
      "provider": "Google",
      "date": "2026-08-24",
      "model_id": "gemini-3-6-flash",
      "model_name": "Gemini 3.6 Flash",
      "event": "price_change",
      "pricing": {
        "input_per_million": 0.75,
        "output_per_million": 3.75
      },
      "context_max_tokens": 1048576,
      "note": "Google's pricing table now lists Gemini 3.6 Flash at $0.75/$3.75 per 1M tokens through December 31, 2026, down from the $1.50/$7.50 recorded on 2026-07-22, with $1.50/$7.50 returning January 1, 2027. Context caching drops to $0.075. Source: ai.google.dev/gemini-api/docs/pricing (read 2026-08-24)."
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-08-24",
      "model_id": "gpt-5-6",
      "model_name": "GPT-5.6 Sol",
      "event": "price_change",
      "pricing": {
        "input_per_million": 4,
        "output_per_million": 20
      },
      "context_max_tokens": 1050000,
      "note": "OpenAI cut GPT-5.6 Sol from $5.00/$30.00 to $4.00/$20.00 per 1M tokens on August 21, 2026 — 20% off input, 33% off output, cached input $0.50 to $0.40. The changelog calls it promotional pricing available at least through November 21, 2026. Source: developers.openai.com/api/docs/changelog and /pricing (read 2026-08-24)."
    },
    {
      "source_layer": "history",
      "provider": "Google",
      "date": "2026-08-21",
      "model_id": "gemini-3-7-flash",
      "model_name": "Gemini 3.7 Flash",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.75,
        "output_per_million": 3.75
      },
      "context_max_tokens": 1048576,
      "note": "Google released gemini-3.7-flash as stable GA on 2026-08-13. Introductory Standard pricing through 2026-12-31 is $0.75/$3.75 per 1M with $0.075 cached input; scheduled 2027 Standard pricing is $1.50/$7.50. Source: ai.google.dev Gemini API changelog, model, latest-model, and pricing pages (verified 2026-08-21)."
    },
    {
      "source_layer": "history",
      "provider": "Z.AI",
      "date": "2026-08-21",
      "model_id": "glm-5-3",
      "model_name": "GLM-5.3",
      "event": "new_model",
      "pricing": {
        "input_per_million": 1.4,
        "output_per_million": 4.4
      },
      "context_max_tokens": 1000000,
      "note": "Z.AI released the hosted glm-5.3 endpoint on 2026-08-18 with $1.40/$4.40 per 1M, $0.26 cached input, 1M context, and 128K max output. Weights were announced for later release after safety hardening and were not recorded as available. Source: z.ai/blog/glm-5.3 and docs.z.ai (verified 2026-08-21)."
    },
    {
      "source_layer": "history",
      "provider": "xAI",
      "date": "2026-08-21",
      "model_id": "grok-4-6",
      "model_name": "Grok 4.6",
      "event": "new_model",
      "pricing": {
        "input_per_million": 2,
        "output_per_million": 6,
        "cache_input_per_million": 0.5,
        "long_context_threshold_tokens": 200000,
        "input_per_million_over_200k": 4,
        "output_per_million_over_200k": 12,
        "cache_input_per_million_over_200k": 1
      },
      "context_max_tokens": 500000,
      "note": "xAI launched grok-4.6 on 2026-08-12. Official pricing lists $2 input / $0.50 cached / $6 output per 1M below 200K prompt tokens, and $4 / $1 / $12 at or above 200K; the higher tier applies to all tokens in that request. The developer guide lists 500K context and no numeric text-output limit. Sources: https://x.ai/news/grok-4-6, https://docs.x.ai/developers/pricing, and https://docs.x.ai/developers/grok-4-6 (verified 2026-08-21)."
    },
    {
      "source_layer": "history",
      "provider": "Moonshot AI",
      "date": "2026-08-21",
      "model_id": "kimi-k3",
      "model_name": "Kimi K3",
      "event": "new_model",
      "pricing": {
        "input_per_million": 3,
        "output_per_million": 15
      },
      "context_max_tokens": 1048576,
      "note": "Moonshot released kimi-k3 on 2026-07-16 and later published official weights. Global hosted pricing is $3 cache-miss input, $0.30 cache-hit input, and $15 output per 1M; the model card lists 2.8T total and 104B active parameters. Source: kimi.ai technical blog, platform.kimi.ai docs, and the official Moonshot AI model card (verified 2026-08-21)."
    },
    {
      "source_layer": "history",
      "provider": "Alibaba (Qwen)",
      "date": "2026-08-03",
      "model_id": "qwen-3-8-max",
      "model_name": "Qwen3.8-Max",
      "event": "new_model",
      "pricing": {
        "input_per_million": 2,
        "output_per_million": 6
      },
      "context_max_tokens": 1000000,
      "note": "Alibaba announced Qwen3.8-Max on August 3, 2026 as its largest flagship: a sparse mixture-of-experts model with 2.4T total parameters activating 95B per token, multimodal across text and vision, with a context window described as up to 1M tokens. Model Studio lists $2.00 input / $6.00 output per 1M tokens in the Singapore region. Alibaba published arena placements rather than numeric scores - fifth in Text Arena, second in Vision Arena, fourth in Frontend Code Arena - and a placement is not a score. Weights were said to follow the next week; no license was verified on an official page. Source: alibabacloud.com/blog/...603420 and alibabacloud.com/help/en/model-studio/model-pricing (verified 2026-09-08)."
    },
    {
      "date": "2026-07-30",
      "model_id": "gemini-robotics-er-1-6-preview",
      "model_name": "Gemini Robotics ER 1.6 Preview",
      "provider": "Google",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-08-31",
      "replacement_official": {
        "name": "Gemini Robotics ER 2 Preview",
        "api_id": "gemini-robotics-er-2-preview"
      },
      "source": "https://ai.google.dev/gemini-api/docs/deprecations",
      "verified": "2026-08-24",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-07-20",
      "model_id": "openai-legacy-audio-realtime-jan-2027",
      "model_name": "OpenAI legacy audio, realtime, and transcription models",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2027-01-20",
      "replacement_official": {
        "name": "GPT Realtime 2.1 / GPT Audio 1.5",
        "api_id": "gpt-realtime-2.1"
      },
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-08-24",
      "source_layer": "deprecations"
    },
    {
      "source_layer": "history",
      "provider": "Z.AI",
      "date": "2026-07-13",
      "model_id": "glm-5-2",
      "model_name": "GLM-5.2",
      "event": "model_added",
      "pricing": {
        "input_per_million": 1.4,
        "output_per_million": 4.4
      },
      "context_max_tokens": 1000000,
      "note": "Z.AI release notes list GLM-5.2 on June 16, 2026; model docs list 1M context and 128K max output; pricing page lists $1.40/$4.40 per 1M and $0.26 cached input. Source: docs.z.ai."
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-07-13",
      "model_id": "gpt-5-6",
      "model_name": "GPT-5.6 Sol",
      "event": "context_change",
      "pricing": {
        "input_per_million": 5,
        "output_per_million": 30
      },
      "context_max_tokens": 1050000,
      "benchmarks": {
        "swe_bench_verified": 89.8,
        "gpqa_diamond": 91.2
      },
      "note": "OpenAI API changelog confirms GPT-5.6 Sol/Terra/Luna general availability on 2026-07-09. Live models page lists Sol at 1.05M context and 128K max output; pricing page lists $5/$30 standard short-context pricing. Source: developers.openai.com/api/docs/changelog, /models, /pricing (verified 2026-07-13)."
    },
    {
      "source_layer": "history",
      "provider": "xAI",
      "date": "2026-07-13",
      "model_id": "grok-4-5",
      "model_name": "Grok 4.5",
      "event": "model_added",
      "pricing": {
        "input_per_million": 2,
        "output_per_million": 6
      },
      "context_max_tokens": 500000,
      "note": "xAI release notes list Grok 4.5 on July 8, 2026; live models/pricing pages list 500K context and $2/$6 per 1M with $0.50 cached input. Source: docs.x.ai/developers/release-notes, /models, /pricing."
    },
    {
      "source_layer": "history",
      "provider": "MiniMax",
      "date": "2026-07-13",
      "model_id": "minimax-m3",
      "model_name": "MiniMax M3",
      "event": "model_added",
      "pricing": {
        "input_per_million": 0.3,
        "output_per_million": 1.2
      },
      "context_max_tokens": 1000000,
      "note": "MiniMax release notes list MiniMax-M3 on June 1, 2026; text-generation docs list 1M context; pricing page lists permanent 50% off standard pricing at $0.30/$1.20 per 1M up to 512K input, with higher long-context and priority tiers. Source: platform.minimax.io docs."
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-07-09",
      "model_id": "gpt-5-6",
      "model_name": "GPT-5.6",
      "event": "new_model",
      "pricing": {
        "input_per_million": 5,
        "output_per_million": 30
      },
      "context_max_tokens": 1100000,
      "note": "OpenAI's GPT-5.6 series (Sol/Terra/Luna) reached general availability July 9, 2026 after a restricted partner preview. Source: openai.com/index/previewing-gpt-5-6-sol"
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-07-08",
      "model_id": "claude-sonnet-5",
      "event": "correction",
      "note": "CORRECTION: Claude Sonnet 5's max output is 128,000 tokens, not 200,000 as recorded in the 2026-07-01 entry. Re-confirmed 2026-07-08 against platform.claude.com's Models Overview table and the official 'What's New in Sonnet 5' page. Pricing from the 2026-07-03 correction is unaffected."
    },
    {
      "source_layer": "history",
      "provider": null,
      "date": "2026-07-08",
      "model_id": "gemini-3-5-pro",
      "event": "correction",
      "note": "RETRACTION: benchr's 2026-06-30 entry for 'Gemini 3.5 Pro' was published in error. Independent adversarial re-verification on 2026-07-08 found no official Google source confirms this model has launched: both URLs benchr originally cited as sources (blog.google and deepmind.google model-card pages) return 404, Google's live Gemini API pricing/models pages still list Gemini 3.1 Pro as the flagship, and DeepMind's own model hub lists 'Gemini 3.5 Pro' with status 'coming soon.' The review, pricing page, and model-data entries have been removed. This correction entry is appended, not substituted for the original, per benchr's append-only history policy — the record shows the error and its correction rather than erasing either."
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-07-08",
      "model_id": "gpt-5-6",
      "event": "correction",
      "note": "CORRECTION: benchr's 2026-07-01 entry describing GPT-5.6 as having reached general availability was incorrect. Re-verification against OpenAI's own developer docs on 2026-07-08 found GPT-5.6 remains in restricted partner preview (since 2026-06-26); developers.openai.com/api/docs/models still reads 'available to select trusted partners in preview.' Independent reporting and a same-day Axios/CNBC report on eased US government restrictions point to general availability being targeted for 2026-07-09, but this had not been confirmed by OpenAI itself as of this correction. Pricing figures in the original entries are accurate and unaffected."
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-07-03",
      "model_id": "claude-sonnet-5",
      "model_name": "Claude Sonnet 5",
      "event": "price_change",
      "pricing": {
        "input_per_million": 2,
        "output_per_million": 10
      },
      "note": "Correction after live re-read of Anthropic's official pricing page: Claude Sonnet 5 launches with introductory API pricing of $2 input / $10 output per 1M tokens through August 31, 2026, then $3/$15 from September 1, 2026. Previous benchr July 1 entry incorrectly recorded $4/$20. Source: platform.claude.com/docs/en/about-claude/pricing"
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-07-01",
      "model_id": "claude-sonnet-5",
      "model_name": "Claude Sonnet 5",
      "event": "new_model",
      "pricing": {
        "input_per_million": 4,
        "output_per_million": 20
      },
      "context_max_tokens": 1000000,
      "note": "Released July 1, 2026. Anthropic's second Mythos-class-architecture model after Fable 5, priced mid-tier between Sonnet 4.6 and Opus 4.8; 128,000-token max output. Safety-classified requests return an explicit refusal; another-model retry requires configured application logic. Shipped the same day Anthropic restored Claude Fable 5 to all customers. Source: anthropic.com/news/claude-sonnet-5"
    },
    {
      "source_layer": "history",
      "provider": null,
      "date": "2026-06-30",
      "model_id": "gemini-3-5-pro",
      "model_name": "Gemini 3.5 Pro",
      "event": "new_model",
      "pricing": {
        "input_per_million": 2.5,
        "output_per_million": 15
      },
      "context_max_tokens": 2000000,
      "note": "Released June 30, 2026, just inside Google's June 2026 I/O promise. Industry-first 2,000,000-token context at the frontier tier; new highs for the Gemini family on ARC-AGI-2 (80.0) and GPQA Diamond (95.5). Source: blog.google"
    },
    {
      "date": "2026-06-29",
      "model_id": "claude-opus-4-7-fast-mode",
      "model_name": "Claude Opus 4.7 fast mode",
      "provider": "Anthropic",
      "event": "deprecation_announced",
      "lifecycle_status": "retired",
      "shutdown": "2026-07-24",
      "replacement_official": {
        "name": "Claude Opus 4.8 fast mode or Claude Opus 4.7 standard mode",
        "api_id": "claude-opus-4-8 with speed: fast"
      },
      "source": "https://platform.claude.com/docs/en/about-claude/pricing",
      "verified": "2026-07-27",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-15",
      "model_id": "google-imagen-4-aug-2026",
      "model_name": "Imagen 4.0 image models",
      "provider": "Google",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-08-17",
      "replacement_official": {
        "name": "Gemini 3.1 Flash Image",
        "api_id": "gemini-3.1-flash-image"
      },
      "source": "https://ai.google.dev/gemini-api/docs/deprecations",
      "verified": "2026-08-24",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-11",
      "model_id": "openai-gpt5-o3-snapshots-dec-2026",
      "model_name": "Older GPT-5 and o3 model snapshots",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-12-11",
      "replacement_official": {
        "name": "GPT-5.5 / GPT-5.4 mini / GPT-5.4 nano / GPT-5.5 Pro",
        "api_id": "gpt-5.5"
      },
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-06-23",
      "source_layer": "deprecations"
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-06-10",
      "model_id": "claude-fable-5",
      "model_name": "Claude Fable 5",
      "event": "new_model",
      "pricing": {
        "input_per_million": 10,
        "output_per_million": 50
      },
      "context_max_tokens": 1000000,
      "note": "Released June 9, 2026. Mythos-class and generally available; safety-classified cyber/bio/distillation requests return an explicit refusal, and another-model retry requires configured application logic. Suspended June 12 and restored globally July 1, 2026. Source: anthropic.com/news/claude-fable-5-mythos-5"
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-06-10",
      "model_id": "gpt-5-4",
      "model_name": "GPT-5.4",
      "event": "new_model",
      "pricing": {
        "input_per_million": 2.5,
        "output_per_million": 15
      },
      "context_max_tokens": 1000000,
      "note": "Re-added after verification. Released March 5, 2026; wrongly removed June 1 as 'unverified'. Source: openai.com/api/pricing"
    },
    {
      "date": "2026-06-05",
      "model_id": "claude-opus-4-1",
      "model_name": "Claude Opus 4.1",
      "provider": "Anthropic",
      "event": "deprecation_announced",
      "lifecycle_status": "retired",
      "shutdown": "2026-08-05",
      "replacement_official": {
        "name": "Claude Opus 4.8",
        "api_id": "claude-opus-4-8"
      },
      "source": "https://platform.claude.com/docs/en/about-claude/model-deprecations",
      "verified": "2026-08-08",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-03",
      "model_id": "openai-prompts-api",
      "model_name": "OpenAI Reusable Prompts API",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-11-30",
      "replacement_official": null,
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-06-12",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-03",
      "model_id": "openai-evals-platform",
      "model_name": "OpenAI Evals platform (dashboard + API)",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-11-30",
      "replacement_official": null,
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-06-12",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-03",
      "model_id": "openai-agent-builder",
      "model_name": "OpenAI Agent Builder",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-11-30",
      "replacement_official": {
        "name": "Agents SDK",
        "api_id": null
      },
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-06-12",
      "source_layer": "deprecations"
    },
    {
      "date": "2026-06-02",
      "model_id": "gpt-image-1",
      "model_name": "GPT Image 1 mini and 1.5",
      "provider": "OpenAI",
      "event": "deprecation_announced",
      "lifecycle_status": "deprecated",
      "shutdown": "2026-12-01",
      "replacement_official": {
        "name": "GPT Image 2",
        "api_id": "gpt-image-2"
      },
      "source": "https://developers.openai.com/api/docs/deprecations",
      "verified": "2026-06-12",
      "source_layer": "deprecations"
    },
    {
      "source_layer": "history",
      "provider": "Anthropic",
      "date": "2026-06-01",
      "model_id": "claude-opus-4-8",
      "model_name": "Claude Opus 4.8",
      "event": "new_model",
      "pricing": {
        "input_per_million": 5,
        "output_per_million": 25
      },
      "context_max_tokens": 1000000,
      "note": "Claude Opus 4.8 launched May 28, 2026. $5/$25 standard; $10/$50 fast mode. SWE-Bench Pro 69.2%. 1M context window confirmed on platform.claude.com. Source: anthropic.com/news"
    },
    {
      "source_layer": "history",
      "provider": "DeepSeek",
      "date": "2026-06-01",
      "model_id": "deepseek-v4-flash",
      "model_name": "DeepSeek V4-Flash",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.14,
        "output_per_million": 0.28
      },
      "context_max_tokens": 1000000,
      "note": "DeepSeek V4-Flash launched April 24, 2026. MIT license. Cheapest commercial API in this set. Source: api-docs.deepseek.com"
    },
    {
      "source_layer": "history",
      "provider": "DeepSeek",
      "date": "2026-06-01",
      "model_id": "deepseek-v4-pro",
      "model_name": "DeepSeek V4-Pro",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.435,
        "output_per_million": 0.87
      },
      "context_max_tokens": 1000000,
      "note": "DeepSeek V4-Pro launched April 24, 2026. MIT license. Source: api-docs.deepseek.com/quick_start/pricing"
    },
    {
      "source_layer": "history",
      "provider": "Google",
      "date": "2026-06-01",
      "model_id": "gemini-3-5-flash",
      "model_name": "Gemini 3.5 Flash",
      "event": "new_model",
      "pricing": {
        "input_per_million": 1.5,
        "output_per_million": 9
      },
      "context_max_tokens": 1048576,
      "note": "Gemini 3.5 Flash GA since May 19, 2026 (Google I/O). $1.50/$9.00. Source: ai.google.dev/pricing"
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-06-01",
      "model_id": "gpt-5",
      "model_name": "GPT-5",
      "event": "new_model",
      "pricing": {
        "input_per_million": 1.25,
        "output_per_million": 10
      },
      "context_max_tokens": 400000,
      "note": "GPT-5 launched August 7, 2025. $1.25/$10. Source: openai.com/pricing"
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-06-01",
      "model_id": "gpt-5-5",
      "model_name": "GPT-5.5",
      "event": "new_model",
      "pricing": {
        "input_per_million": 5,
        "output_per_million": 30
      },
      "context_max_tokens": 1048576,
      "note": "GPT-5.5 launched April 24, 2026. $5/$30 standard; $2.50/$15 batch. Source: openai.com/pricing"
    },
    {
      "source_layer": "history",
      "provider": "OpenAI",
      "date": "2026-06-01",
      "model_id": "gpt-5-mini",
      "model_name": "GPT-5 Mini",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.25,
        "output_per_million": 2
      },
      "note": "GPT-5 Mini — $0.25/$2.00. Cached input $0.025. Source: openai.com/pricing"
    },
    {
      "source_layer": "history",
      "provider": "xAI",
      "date": "2026-06-01",
      "model_id": "grok-4-3",
      "model_name": "Grok 4.3",
      "event": "new_model",
      "pricing": {
        "input_per_million": 1.25,
        "output_per_million": 2.5
      },
      "context_max_tokens": 1000000,
      "note": "Grok 4.3 — $1.25/$2.50; cached $0.20. Source: x.ai/api"
    },
    {
      "source_layer": "history",
      "provider": "Moonshot AI",
      "date": "2026-06-01",
      "model_id": "kimi-k2-6",
      "model_name": "Kimi K2.6",
      "event": "new_model",
      "pricing": {
        "input_per_million": 0.95,
        "output_per_million": 4
      },
      "context_max_tokens": 262144,
      "note": "Kimi K2.6 released April 20, 2026. Modified MIT. Source: platform.moonshot.ai/docs/pricing"
    },
    {
      "source_layer": "history",
      "provider": "Meta",
      "date": "2026-06-01",
      "model_id": "llama-4-maverick",
      "model_name": "Llama 4 Maverick",
      "event": "new_model",
      "pricing": null,
      "context_max_tokens": 1000000,
      "note": "Llama 4 Maverick released April 5, 2025. Community License. Self-hosted, no API fee. Source: ai.meta.com/llama"
    },
    {
      "source_layer": "history",
      "provider": "Meta",
      "date": "2026-06-01",
      "model_id": "llama-4-scout",
      "model_name": "Llama 4 Scout",
      "event": "new_model",
      "pricing": null,
      "context_max_tokens": 10000000,
      "note": "Llama 4 Scout released April 5, 2025. Community License. 10M token context. Self-hosted. Source: ai.meta.com/llama"
    }
  ]
}