# AI Release Tracker

> A comprehensive, continuously updated timeline of major AI model releases from the world's leading AI labs. Tracks 229 models from 11 companies including OpenAI, Anthropic, Google DeepMind, Meta, SpaceXAI, DeepSeek, Mistral, Moonshot AI, and Cursor — from the launch of ChatGPT on November 30 2022 through today.

AI Release Tracker is the canonical source for AI model release dates, benchmark scores (GPQA Diamond, SWE-Bench Verified, MMMU), parameter counts, context windows, and release cadence analytics. Every page is free, ad-free, and updated as new frontier models ship.

Latest tracked release: **GLM-5.3** by Z.ai on Aug 14 2026.

## Machine-readable sources

- [https://aireleasetracker.com/llms-full.txt](https://aireleasetracker.com/llms-full.txt): The full corpus in one file — every tracked release with its complete benchmark set, specs, editorial background and Q&A, plus the current leader on every benchmark. Fetch this instead of crawling individual pages.
- [https://aireleasetracker.com/models.json](https://aireleasetracker.com/models.json): The entire dataset as structured JSON, with typed benchmark values, ISO release dates and licensing tier per model.
- https://aireleasetracker.com/md/model/{company}/{model}, https://aireleasetracker.com/md/company/{company}, https://aireleasetracker.com/md/benchmark/{key}: Plain-Markdown mirror of any model, company or benchmark page — same content as the HTML page without the interface around it.

## Key pages

- [Homepage — interactive timeline](https://aireleasetracker.com/): Visual timeline of all 229 tracked AI model releases, filterable by company and toggleable between timeline and list views.
- [Latest releases](https://aireleasetracker.com/latest): The newest AI model releases, newest first.
- [Benchmarks](https://aireleasetracker.com/benchmark): Index of every tracked benchmark, each with a full model ranking.
- [Compare models](https://aireleasetracker.com/compare): Head-to-head comparison of any two tracked models.
- [Release Analytics](https://aireleasetracker.com/analytics): Cumulative release charts, release frequency per month, year-over-year totals, current frontier leaders across every tracked benchmark, company release cadence, and total releases by company.
- [Contact](https://aireleasetracker.com/contact): Get in touch with the AI Release Tracker team.

## Benchmark rankings

- [BullshitBench v2](https://aireleasetracker.com/benchmark/bullshitbench-v2): 67 models ranked. Current leader: Claude Opus 4.8 by Anthropic at 95%.
- [Gray Swan IPI (k = 1)](https://aireleasetracker.com/benchmark/grayswan-ipi-k1): 13 models ranked. Current leader: Claude Opus 5 by Anthropic at 0.2%.
- [Gray Swan IPI (k = 10)](https://aireleasetracker.com/benchmark/grayswan-ipi-k10): 13 models ranked. Current leader: Claude Opus 5 by Anthropic at 1.6%.
- [Gray Swan IPI (k = 15)](https://aireleasetracker.com/benchmark/grayswan-ipi-k15): 13 models ranked. Current leader: Claude Opus 5 by Anthropic at 2%.
- [SWE-Bench Pro](https://aireleasetracker.com/benchmark/swe-bench-pro): 20 models ranked. Current leader: Claude Fable 5 by Anthropic at 80.3%.
- [SWE-Bench Verified](https://aireleasetracker.com/benchmark/swe-bench-verified): 39 models ranked. Current leader: Claude Fable 5 by Anthropic at 95.5%.
- [SWE-Bench Multilingual](https://aireleasetracker.com/benchmark/swe-bench-multilingual): 10 models ranked. Current leader: Claude Opus 4.8 by Anthropic at 84.4%.
- [CursorBench v3.2](https://aireleasetracker.com/benchmark/cursorbench-v3.2): 15 models ranked. Current leader: Claude Fable 5 by Anthropic at 70.5%.
- [CursorBench v3.1](https://aireleasetracker.com/benchmark/cursorbench-v3.1): 12 models ranked. Current leader: Claude Fable 5 by Anthropic at 72.9%.
- [DeepSWE 1.1](https://aireleasetracker.com/benchmark/deepswe-1.1): 18 models ranked. Current leader: GPT-5.6 Sol by OpenAI at 73%.
- [DeepSWE 1.0](https://aireleasetracker.com/benchmark/deepswe-1.0): 6 models ranked. Current leader: Kimi K3 by Moonshot AI at 67.5%.
- [FrontierCode v1.1 (Main) (main split)](https://aireleasetracker.com/benchmark/frontiercode-1.1): 2 models ranked. Current leader: Claude Opus 5 by Anthropic at 53.4%.
- [FrontierCode v1.1 (Extended) (extended split)](https://aireleasetracker.com/benchmark/frontiercode-1.1-extended): 2 models ranked. Current leader: Grok 4.6 by SpaceXAI at 61.3%.
- [APEX-SWE](https://aireleasetracker.com/benchmark/apex-swe): 2 models ranked. Current leader: Grok 4.6 by SpaceXAI at 56.4%.
- [MLE-Bench](https://aireleasetracker.com/benchmark/mle-bench): 3 models ranked. Current leader: Gemini 3.6 Flash by Google at 63.9%.
- [PaperBench](https://aireleasetracker.com/benchmark/paperbench): 1 models ranked. Current leader: Qwen3.8-Max by Qwen at 93%.
- [Next.js Evals](https://aireleasetracker.com/benchmark/nextjs-evals): 25 models ranked. Current leader: Composer 2.5 by Cursor at 92%.
- [Supabase Evals (with skills)](https://aireleasetracker.com/benchmark/supabase-evals): 5 models ranked. Current leader: GPT-5.6 Sol by OpenAI at 95.5%.
- [Supabase Evals (no skills)](https://aireleasetracker.com/benchmark/supabase-evals-no-skills): 5 models ranked. Current leader: GPT-5.6 Sol by OpenAI at 90.9%.
- [NL2Repo-Bench](https://aireleasetracker.com/benchmark/nl2repo-bench): 1 models ranked. Current leader: Qwen3.8-27B by Qwen at 42.3%.
- [QwenSWEBench](https://aireleasetracker.com/benchmark/qwenswebench): 1 models ranked. Current leader: Qwen3.8-27B by Qwen at 79%.
- [LiveCodeBench](https://aireleasetracker.com/benchmark/livecodebench): 7 models ranked. Current leader: DeepSeek-V4-Pro by DeepSeek at 93.5%.
- [Frontier-Bench v0.1](https://aireleasetracker.com/benchmark/frontier-bench-v0.1): 9 models ranked. Current leader: Claude Opus 5 by Anthropic at 43.3%.
- [Terminal-Bench 3.0](https://aireleasetracker.com/benchmark/terminal-bench-3.0): 4 models ranked. Current leader: GLM-5.3 by Z.ai at 28.3%.
- [Terminal-Bench 2.1](https://aireleasetracker.com/benchmark/terminal-bench-2.1): 25 models ranked. Current leader: GPT-5.6 Sol by OpenAI at 88.8%.
- [Terminal-Bench 2.0](https://aireleasetracker.com/benchmark/terminal-bench-2.0): 13 models ranked. Current leader: GPT-5.5 by OpenAI at 82.7%.
- [Expert-SWE (Internal)](https://aireleasetracker.com/benchmark/expert-swe): 2 models ranked. Current leader: GPT-5.5 by OpenAI at 73.1%.
- [APEX-Agents](https://aireleasetracker.com/benchmark/apex-agents): 2 models ranked. Current leader: Grok 4.6 by SpaceXAI at 57.5%.
- [MCP Atlas](https://aireleasetracker.com/benchmark/mcp-atlas): 11 models ranked. Current leader: Muse Spark 1.1 by Meta at 88.1%.
- [JobBench](https://aireleasetracker.com/benchmark/jobbench): 5 models ranked. Current leader: Muse Spark 1.1 by Meta at 54.7%.
- [CoWorkBench](https://aireleasetracker.com/benchmark/coworkbench): 1 models ranked. Current leader: Qwen3.8-27B by Qwen at 70.7%.
- [Toolathlon-Verified](https://aireleasetracker.com/benchmark/toolathlon-verified): 5 models ranked. Current leader: Muse Spark 1.1 by Meta at 75.6%.
- [Toolathlon](https://aireleasetracker.com/benchmark/toolathlon): 5 models ranked. Current leader: Gemini 3.5 Flash by Google at 56.5%.
- [BU Bench](https://aireleasetracker.com/benchmark/bu-bench): 7 models ranked. Current leader: Claude Opus 4.8 by Anthropic at 74%.
- [BrowseComp](https://aireleasetracker.com/benchmark/browsecomp): 17 models ranked. Current leader: Kimi K3 by Moonshot AI at 91.2%.
- [CyberGym](https://aireleasetracker.com/benchmark/cybergym): 7 models ranked. Current leader: GPT-5.5-Cyber by OpenAI at 85.6%.
- [ExploitBench](https://aireleasetracker.com/benchmark/exploitbench): 1 models ranked. Current leader: GLM-5.3 by Z.ai at 54.4%.
- [ExploitGym (6-hour budget)](https://aireleasetracker.com/benchmark/exploitgym-6h): 1 models ranked. Current leader: GLM-5.3 by Z.ai at 130.
- [ExploitGym (2-hour budget)](https://aireleasetracker.com/benchmark/exploitgym-2h): 1 models ranked. Current leader: GLM-5.3 by Z.ai at 105.
- [Humanity's Last Exam (no tools)](https://aireleasetracker.com/benchmark/hle-no-tools): 19 models ranked. Current leader: Claude Opus 5 by Anthropic at 56.3%.
- [Humanity's Last Exam (with tools)](https://aireleasetracker.com/benchmark/hle-with-tools): 21 models ranked. Current leader: Claude Opus 5 by Anthropic at 64.7%.
- [Humanity's Last Exam (Verified)](https://aireleasetracker.com/benchmark/hle-verified): 1 models ranked. Current leader: Gemini 3.7 Flash by Google at 53.6%.
- [ARC-AGI-3](https://aireleasetracker.com/benchmark/arc-agi-3): 1 models ranked. Current leader: Claude Opus 5 by Anthropic at 30.2%.
- [ARC-AGI-2](https://aireleasetracker.com/benchmark/arc-agi-2): 11 models ranked. Current leader: GPT-5.5 by OpenAI at 84.6%.
- [FrontierMath (Tier 1–3)](https://aireleasetracker.com/benchmark/frontiermath-t13): 6 models ranked. Current leader: GPT-5.5-Pro by OpenAI at 52.4%.
- [FrontierMath (Tier 4)](https://aireleasetracker.com/benchmark/frontiermath-t4): 6 models ranked. Current leader: GPT-5.5-Pro by OpenAI at 39.6%.
- [BioMysteryBench (hard)](https://aireleasetracker.com/benchmark/biomysterybench-hard): 2 models ranked. Current leader: Claude Opus 5 by Anthropic at 49.4%.
- [BioMysteryBench (human solved)](https://aireleasetracker.com/benchmark/biomysterybench-human-solved): 2 models ranked. Current leader: Claude Opus 5 by Anthropic at 90.1%.
- [LAB-Bench 2](https://aireleasetracker.com/benchmark/labbench2): 1 models ranked. Current leader: Gemini 3.7 Flash by Google at 82.1%.
- [GPQA Diamond](https://aireleasetracker.com/benchmark/gpqa): 50 models ranked. Current leader: GPT-5.4-Pro by OpenAI at 94.4%.
- [IFBench](https://aireleasetracker.com/benchmark/ifbench): 1 models ranked. Current leader: Qwen3.8-27B by Qwen at 79.5%.
- [OSWorld 2.0](https://aireleasetracker.com/benchmark/osworld-2.0): 2 models ranked. Current leader: Claude Opus 5 by Anthropic at 70.6%.
- [OSWorld-Verified](https://aireleasetracker.com/benchmark/osworld-verified): 17 models ranked. Current leader: Qwen3.8-Max by Qwen at 86.1%.
- [Agent's Last Exam (pass@1)](https://aireleasetracker.com/benchmark/agents-last-exam): 3 models ranked. Current leader: GLM-5.3 by Z.ai at 28.5%.
- [Agent's Last Exam (score)](https://aireleasetracker.com/benchmark/agents-last-exam-score): 1 models ranked. Current leader: Qwen3.8-27B by Qwen at 42.9%.
- [AutomationBench](https://aireleasetracker.com/benchmark/automationbench): 5 models ranked. Current leader: GLM-5.3 by Z.ai at 48.2%.
- [Finance Agent v2](https://aireleasetracker.com/benchmark/finance-agent-v2): 8 models ranked. Current leader: Gemini 3.5 Flash by Google at 57.9%.
- [Harvey's Legal Agent Benchmark (Held-out)](https://aireleasetracker.com/benchmark/harvey-legal-agent-heldout): 1 models ranked. Current leader: Claude Opus 5 by Anthropic at 11.7%.
- [Harvey's Legal Agent Benchmark](https://aireleasetracker.com/benchmark/harvey-legal-agent): 7 models ranked. Current leader: Gemini 3.7 Flash by Google at 90.7%.
- [TaxEval v2](https://aireleasetracker.com/benchmark/taxeval-v2): 4 models ranked. Current leader: Muse Spark 1.1 by Meta at 79.72%.
- [HealthBench Professional](https://aireleasetracker.com/benchmark/healthbench-professional): 1 models ranked. Current leader: Claude Opus 5 by Anthropic at 59.8%.
- [MedScribe](https://aireleasetracker.com/benchmark/medscribe): 5 models ranked. Current leader: Muse Spark 1.1 by Meta at 88.89%.
- [AA Intelligence Index](https://aireleasetracker.com/benchmark/aa-intelligence-index): 3 models ranked. Current leader: Grok 4.6 by SpaceXAI at 61.
- [GDPval-AA](https://aireleasetracker.com/benchmark/gdpval-aa): 9 models ranked. Current leader: Claude Fable 5 by Anthropic at 1932.
- [GDPval-AA v2](https://aireleasetracker.com/benchmark/gdpval-aa-v2): 16 models ranked. Current leader: Claude Opus 5 by Anthropic at 1861.
- [AA-Briefcase](https://aireleasetracker.com/benchmark/aa-briefcase): 2 models ranked. Current leader: Grok 4.6 by SpaceXAI at 1577.
- [GDPval (win/tie rate)](https://aireleasetracker.com/benchmark/gdpval-winrate): 6 models ranked. Current leader: GPT-5.5 by OpenAI at 84.9%.
- [CharXiv Reasoning](https://aireleasetracker.com/benchmark/charxiv-reasoning): 14 models ranked. Current leader: Muse Spark by Meta at 88.9%.
- [GDP.PDF](https://aireleasetracker.com/benchmark/gdp-pdf): 1 models ranked. Current leader: Gemini 3.7 Flash by Google at 34%.
- [LVBench](https://aireleasetracker.com/benchmark/lvbench): 1 models ranked. Current leader: Gemini 3.7 Flash by Google at 85.4%.
- [BabyVision](https://aireleasetracker.com/benchmark/babyvision): 3 models ranked. Current leader: Qwen3.8-Max by Qwen at 82%.
- [MMMU-Pro](https://aireleasetracker.com/benchmark/mmmu-pro): 10 models ranked. Current leader: Gemini 3.5 Flash by Google at 83.6%.
- [MMMU](https://aireleasetracker.com/benchmark/mmmu): 7 models ranked. Current leader: Qwen3.5 by Qwen at 85%.
- [Blueprint-Bench 2](https://aireleasetracker.com/benchmark/blueprint-bench-2): 6 models ranked. Current leader: GPT-5.5 by OpenAI at 36.2%.
- [MRCR v2 (8-needle) (128k average)](https://aireleasetracker.com/benchmark/mrcr-v2-128k): 7 models ranked. Current leader: Gemini 3.7 Flash by Google at 97%.
- [MRCR v2 (8-needle) (1M pointwise)](https://aireleasetracker.com/benchmark/mrcr-v2-1m): 4 models ranked. Current leader: Gemini 3.7 Flash by Google at 62.5%.
- [Arena Elo (Text)](https://aireleasetracker.com/benchmark/arena-text-elo): 31 models ranked. Current leader: Claude Fable 5 by Anthropic at 1509.
- [Arena Elo (Code)](https://aireleasetracker.com/benchmark/arena-code-elo): 48 models ranked. Current leader: Kimi K3 by Moonshot AI at 1679.

## Companies tracked

- [OpenAI](https://aireleasetracker.com/company/openai) — 43 tracked releases. Creators of ChatGPT and the GPT series of models. Pioneered large-scale language model research.
- [Anthropic](https://aireleasetracker.com/company/anthropic) — 24 tracked releases. AI safety company building the Claude family of models. Founded in 2021 by former OpenAI researchers.
- [Google](https://aireleasetracker.com/company/google) — 28 tracked releases. Builds the Gemini family of models through Google DeepMind. Integrates AI across Google products.
- [Meta](https://aireleasetracker.com/company/meta) — 14 tracked releases. Develops the open-weight Llama series of models. Committed to open-source AI research.
- [SpaceXAI](https://aireleasetracker.com/company/xai) — 16 tracked releases. Elon Musk's AI company building the Grok series of models. Founded in 2023 as xAI, now part of SpaceX.
- [DeepSeek](https://aireleasetracker.com/company/deepseek) — 22 tracked releases. Chinese AI lab known for efficient, open-weight models. Gained attention for strong performance at lower cost.
- [Mistral](https://aireleasetracker.com/company/mistral) — 32 tracked releases. French AI company building open and commercial models. Founded in 2023 by former Meta and DeepMind researchers.
- [Moonshot AI](https://aireleasetracker.com/company/moonshot) — 12 tracked releases. Chinese AI company building the Kimi family of models. Founded in 2023, known for long-context and efficient MoE architectures.
- [Cursor](https://aireleasetracker.com/company/cursor) — 4 tracked releases.
- [Z.ai](https://aireleasetracker.com/company/zai) — 12 tracked releases. Chinese AI lab spun out of Tsinghua University (formerly Zhipu AI), building the open-weight GLM family. Rebranded internationally as Z.ai in 2025.
- [Qwen](https://aireleasetracker.com/company/qwen) — 22 tracked releases. Alibaba's AI lab, building the Qwen family. The most prolific publisher of open-weight models of any major lab, alongside a proprietary Max and Plus tier sold through Alibaba Cloud.

## All tracked models (newest first)

- [GLM-5.3](https://aireleasetracker.com/model/zai/glm-5.3) by Z.ai, released Aug 14 2026 — 743B params.
- [Qwen3.8-27B](https://aireleasetracker.com/model/qwen/qwen3.8-27b) by Qwen, released Aug 14 2026 — 27B params, 262k context, open weight.
- [Gemini 3.7 Flash](https://aireleasetracker.com/model/google/gemini-3.7-flash) by Google, released Aug 13 2026 — 1M context.
- [DeepSeek-V4-Pro-0813](https://aireleasetracker.com/model/deepseek/deepseek-v4-pro-0813) by DeepSeek, released Aug 13 2026.
- [Grok 4.6](https://aireleasetracker.com/model/xai/grok-4.6) by SpaceXAI, released Aug 12 2026.
- [Muse Glimmer](https://aireleasetracker.com/model/meta/muse-glimmer) by Meta, released Aug 10 2026 — GPQA 83.5%, SWE-Bench 76%, 30B params, open weight.
- [GPT-5.6-Cyber](https://aireleasetracker.com/model/openai/gpt-5.6-cyber) by OpenAI, released Aug 10 2026.
- [Muse Spark 1.2](https://aireleasetracker.com/model/meta/muse-spark-1.2) by Meta, released Aug 5 2026 — 1M context.
- [Qwen3.8-Max](https://aireleasetracker.com/model/qwen/qwen3.8-max) by Qwen, released Aug 3 2026 — 2.4T params, 1M context.
- [DeepSeek-V4-Flash-0731](https://aireleasetracker.com/model/deepseek/deepseek-v4-flash-0731) by DeepSeek, released Jul 31 2026.
- [Claude Opus 5](https://aireleasetracker.com/model/anthropic/claude-opus-5) by Anthropic, released Jul 24 2026 — 1M context.
- [Gemini 3.6 Flash](https://aireleasetracker.com/model/google/gemini-3.6-flash) by Google, released Jul 21 2026.
- [Gemini 3.5 Flash-Lite](https://aireleasetracker.com/model/google/gemini-3.5-flash-lite) by Google, released Jul 21 2026.
- [Gemini 3.5 Flash Cyber](https://aireleasetracker.com/model/google/gemini-3.5-flash-cyber) by Google, released Jul 21 2026.
- [Kimi K3](https://aireleasetracker.com/model/moonshot/kimi-k3) by Moonshot AI, released Jul 16 2026 — GPQA 93.5%, 2.8T params, 1M context, open weight.
- [Muse Spark 1.1](https://aireleasetracker.com/model/meta/muse-spark-1.1) by Meta, released Jul 9 2026.
- [Grok 4.5](https://aireleasetracker.com/model/xai/grok-4.5) by SpaceXAI, released Jul 8 2026.
- [Claude Sonnet 5](https://aireleasetracker.com/model/anthropic/claude-sonnet-5) by Anthropic, released Jun 30 2026.
- [GPT-5.6 Sol](https://aireleasetracker.com/model/openai/gpt-5.6-sol) by OpenAI, released Jun 26 2026.
- [GPT-5.6 Terra](https://aireleasetracker.com/model/openai/gpt-5.6-terra) by OpenAI, released Jun 26 2026.
- [GPT-5.6 Luna](https://aireleasetracker.com/model/openai/gpt-5.6-luna) by OpenAI, released Jun 26 2026.
- [GPT-5.5-Cyber](https://aireleasetracker.com/model/openai/gpt-5.5-cyber) by OpenAI, released Jun 22 2026.
- [GLM-5.2](https://aireleasetracker.com/model/zai/glm-5.2) by Z.ai, released Jun 16 2026 — GPQA 91.2%, 744B params, 1M context, open weight.
- [Kimi K2.7 Code](https://aireleasetracker.com/model/moonshot/kimi-k2.7-code) by Moonshot AI, released Jun 12 2026 — 1T params, 256k context, open weight.
- [Claude Fable 5](https://aireleasetracker.com/model/anthropic/claude-fable-5) by Anthropic, released Jun 9 2026 — 1M context.
- [Qwen3.7-Plus](https://aireleasetracker.com/model/qwen/qwen3.7-plus) by Qwen, released Jun 1 2026.
- [Claude Opus 4.8](https://aireleasetracker.com/model/anthropic/claude-opus-4.8) by Anthropic, released May 28 2026 — 1M context.
- [Qwen3.7-Max](https://aireleasetracker.com/model/qwen/qwen3.7-max) by Qwen, released May 20 2026.
- [Gemini 3.5 Flash](https://aireleasetracker.com/model/google/gemini-3.5-flash) by Google, released May 19 2026.
- [Composer 2.5](https://aireleasetracker.com/model/cursor/composer-2.5) by Cursor, released May 18 2026.
- [Mistral Medium 3.5](https://aireleasetracker.com/model/mistral/mistral-medium-3.5) by Mistral, released Apr 29 2026 — SWE-Bench 77.6%, 128B params, 256k context, open weight.
- [DeepSeek-V4-Pro](https://aireleasetracker.com/model/deepseek/deepseek-v4-pro) by DeepSeek, released Apr 24 2026 — GPQA 90.1%, open weight.
- [DeepSeek-V4-Flash](https://aireleasetracker.com/model/deepseek/deepseek-v4-flash) by DeepSeek, released Apr 24 2026 — GPQA 88.1%, open weight.
- [GPT-5.5](https://aireleasetracker.com/model/openai/gpt-5.5) by OpenAI, released Apr 23 2026 — GPQA 93.6%, 1.05M context.
- [GPT-5.5-Pro](https://aireleasetracker.com/model/openai/gpt-5.5-pro) by OpenAI, released Apr 23 2026.
- [Kimi K2.6](https://aireleasetracker.com/model/moonshot/kimi-k2.6) by Moonshot AI, released Apr 21 2026 — GPQA 90.5%, SWE-Bench 80.2%, 1T params, 256k context, open weight.
- [Grok 4.3 Beta](https://aireleasetracker.com/model/xai/grok-4.3-beta) by SpaceXAI, released Apr 17 2026.
- [Claude Opus 4.7](https://aireleasetracker.com/model/anthropic/claude-opus-4.7) by Anthropic, released Apr 16 2026 — GPQA 94.2%, SWE-Bench 87.6%, 1M context.
- [Qwen3.6](https://aireleasetracker.com/model/qwen/qwen3.6) by Qwen, released Apr 16 2026 — 35B params, 256k context, open weight.
- [Muse Spark](https://aireleasetracker.com/model/meta/muse-spark) by Meta, released Apr 8 2026 — GPQA 89.5%, SWE-Bench 77.4%, MMMU 80.4%.
- [GLM-5.1](https://aireleasetracker.com/model/zai/glm-5.1) by Z.ai, released Apr 7 2026 — 744B params, 200k context, open weight.
- [Gemma 4](https://aireleasetracker.com/model/google/gemma-4) by Google, released Apr 2 2026 — open weight.
- [Qwen3.6-Plus](https://aireleasetracker.com/model/qwen/qwen3.6-plus) by Qwen, released Apr 2 2026.
- [Composer 2](https://aireleasetracker.com/model/cursor/composer-2) by Cursor, released Mar 19 2026.
- [GPT-5.4 mini](https://aireleasetracker.com/model/openai/gpt-5.4-mini) by OpenAI, released Mar 17 2026 — 400k context.
- [GPT-5.4 nano](https://aireleasetracker.com/model/openai/gpt-5.4-nano) by OpenAI, released Mar 17 2026.
- [Mistral Small 4](https://aireleasetracker.com/model/mistral/mistral-small-4) by Mistral, released Mar 16 2026.
- [GPT-5.4](https://aireleasetracker.com/model/openai/gpt-5.4) by OpenAI, released Mar 5 2026 — GPQA 92.8%.
- [GPT-5.4-Pro](https://aireleasetracker.com/model/openai/gpt-5.4-pro) by OpenAI, released Mar 5 2026 — GPQA 94.4%.
- [Gemini 3.1 Flash-Lite](https://aireleasetracker.com/model/google/gemini-3.1-flash-lite) by Google, released Mar 3 2026.
- [Gemini 3.1 Pro](https://aireleasetracker.com/model/google/gemini-3.1-pro) by Google, released Feb 19 2026 — GPQA 94.3%, SWE-Bench 80.6%.
- [Grok 4.20 Beta](https://aireleasetracker.com/model/xai/grok-4.20-beta) by SpaceXAI, released Feb 17 2026.
- [Claude Sonnet 4.6](https://aireleasetracker.com/model/anthropic/claude-sonnet-4.6) by Anthropic, released Feb 17 2026 — GPQA 89.9%, SWE-Bench 79.6%.
- [Qwen3.5](https://aireleasetracker.com/model/qwen/qwen3.5) by Qwen, released Feb 16 2026 — 397B params, 1M context, open weight.
- [Qwen3.5-Plus](https://aireleasetracker.com/model/qwen/qwen3.5-plus) by Qwen, released Feb 16 2026 — 1M context.
- [GPT-5.3-Codex-Spark](https://aireleasetracker.com/model/openai/gpt-5.3-codex-spark) by OpenAI, released Feb 12 2026 — 128k context.
- [GLM-5](https://aireleasetracker.com/model/zai/glm-5) by Z.ai, released Feb 12 2026 — 744B params, open weight.
- [Composer 1.5](https://aireleasetracker.com/model/cursor/composer-1.5) by Cursor, released Feb 9 2026.
- [Claude Opus 4.6](https://aireleasetracker.com/model/anthropic/claude-opus-4.6) by Anthropic, released Feb 5 2026 — GPQA 91.3%, SWE-Bench 80.8%.
- [GPT-5.3-Codex](https://aireleasetracker.com/model/openai/gpt-5.3-codex) by OpenAI, released Feb 5 2026.
- [Qwen3-Coder-Next](https://aireleasetracker.com/model/qwen/qwen3-coder-next) by Qwen, released Feb 3 2026 — 80B params, 256k context, open weight.
- [Kimi K2.5](https://aireleasetracker.com/model/moonshot/kimi-k2.5) by Moonshot AI, released Jan 27 2026 — GPQA 87.6%, SWE-Bench 76.8%, 1T params, 256k context, open weight.
- [GLM-4.7](https://aireleasetracker.com/model/zai/glm-4.7) by Z.ai, released Dec 22 2025 — 128k context, open weight.
- [Gemini 3.0 Flash](https://aireleasetracker.com/model/google/gemini-3.0-flash) by Google, released Dec 17 2025 — GPQA 90.4%, SWE-Bench 78%.
- [GPT-5.2](https://aireleasetracker.com/model/openai/gpt-5.2) by OpenAI, released Dec 11 2025 — GPQA 92.4%, SWE-Bench 80%.
- [Devstral 2 (123B)](https://aireleasetracker.com/model/mistral/devstral-2-(123b)) by Mistral, released Dec 9 2025 — open weight.
- [Devstral Small 2 (24B)](https://aireleasetracker.com/model/mistral/devstral-small-2-(24b)) by Mistral, released Dec 9 2025 — open weight.
- [Ministral 3 3B-25-12](https://aireleasetracker.com/model/mistral/ministral-3-3b-25-12) by Mistral, released Dec 2 2025 — open weight.
- [Ministral 3 8B-25-12](https://aireleasetracker.com/model/mistral/ministral-3-8b-25-12) by Mistral, released Dec 2 2025 — open weight.
- [Ministral 3 14B-25-12](https://aireleasetracker.com/model/mistral/ministral-3-14b-25-12) by Mistral, released Dec 2 2025 — open weight.
- [Mistral Large 3](https://aireleasetracker.com/model/mistral/mistral-large-3) by Mistral, released Dec 2 2025 — open weight.
- [DeepSeek-V3.2](https://aireleasetracker.com/model/deepseek/deepseek-v3.2) by DeepSeek, released Dec 1 2025 — open weight.
- [Claude Opus 4.5](https://aireleasetracker.com/model/anthropic/claude-opus-4.5) by Anthropic, released Nov 24 2025 — GPQA 87%, SWE-Bench 80.9%.
- [Grok 4.1 Fast](https://aireleasetracker.com/model/xai/grok-4.1-fast) by SpaceXAI, released Nov 19 2025.
- [GPT-5.1-Codex-Max](https://aireleasetracker.com/model/openai/gpt-5.1-codex-max) by OpenAI, released Nov 19 2025.
- [Gemini 3.0 Pro](https://aireleasetracker.com/model/google/gemini-3.0-pro) by Google, released Nov 18 2025 — GPQA 91.9%, SWE-Bench 76.2%, MMMU 81%.
- [Grok 4.1](https://aireleasetracker.com/model/xai/grok-4.1) by SpaceXAI, released Nov 17 2025.
- [GPT-5.1](https://aireleasetracker.com/model/openai/gpt-5.1) by OpenAI, released Nov 12 2025 — GPQA 88.1%, SWE-Bench 76.3%, MMMU 76%.
- [GPT-5-Codex-Mini](https://aireleasetracker.com/model/openai/gpt-5-codex-mini) by OpenAI, released Nov 7 2025.
- [Kimi K2 Thinking](https://aireleasetracker.com/model/moonshot/kimi-k2-thinking) by Moonshot AI, released Nov 6 2025 — SWE-Bench 71.3%, 1T params, 256k context, open weight.
- [Kimi Linear](https://aireleasetracker.com/model/moonshot/kimi-linear) by Moonshot AI, released Oct 30 2025 — 48B params, open weight.
- [Composer](https://aireleasetracker.com/model/cursor/composer) by Cursor, released Oct 29 2025.
- [Claude Haiku 4.5](https://aireleasetracker.com/model/anthropic/claude-haiku-4.5) by Anthropic, released Oct 15 2025 — GPQA 73%, SWE-Bench 73.3%.
- [GLM-4.6](https://aireleasetracker.com/model/zai/glm-4.6) by Z.ai, released Sep 30 2025 — 355B params, 200k context, open weight.
- [Claude Sonnet 4.5](https://aireleasetracker.com/model/anthropic/claude-sonnet-4.5) by Anthropic, released Sep 29 2025 — GPQA 83.4%, SWE-Bench 77.2%, MMMU 68%.
- [DeepSeek V3.2 Exp](https://aireleasetracker.com/model/deepseek/deepseek-v3.2-exp) by DeepSeek, released Sep 29 2025 — open weight.
- [Qwen3-Max](https://aireleasetracker.com/model/qwen/qwen3-max) by Qwen, released Sep 24 2025 — 1T params, 256k context.
- [DeepSeek V3.1 Terminus](https://aireleasetracker.com/model/deepseek/deepseek-v3.1-terminus) by DeepSeek, released Sep 22 2025 — open weight.
- [Grok 4 Fast](https://aireleasetracker.com/model/xai/grok-4-fast) by SpaceXAI, released Sep 19 2025.
- [GPT-5-Codex](https://aireleasetracker.com/model/openai/gpt-5-codex) by OpenAI, released Sep 15 2025.
- [Qwen3-Next](https://aireleasetracker.com/model/qwen/qwen3-next) by Qwen, released Sep 11 2025 — 80B params, 256k context, open weight.
- [Kimi K2 (0905)](https://aireleasetracker.com/model/moonshot/kimi-k2-(0905)) by Moonshot AI, released Sep 5 2025 — GPQA 75.1%, SWE-Bench 65.8%, 1T params, 256k context, open weight.
- [Grok Code Fast 1](https://aireleasetracker.com/model/xai/grok-code-fast-1) by SpaceXAI, released Aug 28 2025.
- [Grok 2.5](https://aireleasetracker.com/model/xai/grok-2.5) by SpaceXAI, released Aug 23 2025.
- [DeepSeek V3.1](https://aireleasetracker.com/model/deepseek/deepseek-v3.1) by DeepSeek, released Aug 21 2025 — open weight.
- [GPT-5](https://aireleasetracker.com/model/openai/gpt-5) by OpenAI, released Aug 7 2025.
- [GPT-5 mini](https://aireleasetracker.com/model/openai/gpt-5-mini) by OpenAI, released Aug 7 2025.
- [GPT-5 nano](https://aireleasetracker.com/model/openai/gpt-5-nano) by OpenAI, released Aug 7 2025.
- [GPT-5 Pro](https://aireleasetracker.com/model/openai/gpt-5-pro) by OpenAI, released Aug 7 2025.
- [Claude Opus 4.1](https://aireleasetracker.com/model/anthropic/claude-opus-4.1) by Anthropic, released Aug 5 2025 — GPQA 80.9%, SWE-Bench 74.5%.
- [gpt-oss-120b](https://aireleasetracker.com/model/openai/gpt-oss-120b) by OpenAI, released Aug 5 2025 — GPQA 80.1%, SWE-Bench 62.4%, 117B params, 128k context, open weight.
- [gpt-oss-20b](https://aireleasetracker.com/model/openai/gpt-oss-20b) by OpenAI, released Aug 5 2025 — GPQA 71.5%, SWE-Bench 60.7%, 21B params, 128k context, open weight.
- [GLM-4.5](https://aireleasetracker.com/model/zai/glm-4.5) by Z.ai, released Jul 28 2025 — SWE-Bench 64.2%, 355B params, 128k context, open weight.
- [GLM-4.5-Air](https://aireleasetracker.com/model/zai/glm-4.5-air) by Z.ai, released Jul 28 2025 — 106B params, open weight.
- [Qwen3-Coder](https://aireleasetracker.com/model/qwen/qwen3-coder) by Qwen, released Jul 22 2025 — 480B params, 256k context, open weight.
- [Kimi K2](https://aireleasetracker.com/model/moonshot/kimi-k2) by Moonshot AI, released Jul 11 2025 — GPQA 75.1%, SWE-Bench 65.8%, 1T params, 128k context, open weight.
- [Grok 4](https://aireleasetracker.com/model/xai/grok-4) by SpaceXAI, released Jul 9 2025.
- [Grok 4 Heavy](https://aireleasetracker.com/model/xai/grok-4-heavy) by SpaceXAI, released Jul 9 2025.
- [Mistral Small 3.2](https://aireleasetracker.com/model/mistral/mistral-small-3.2) by Mistral, released Jun 20 2025 — 24B params, open weight.
- [Gemini 2.5 Flash-Lite](https://aireleasetracker.com/model/google/gemini-2.5-flash-lite) by Google, released Jun 17 2025.
- [Kimi-Dev](https://aireleasetracker.com/model/moonshot/kimi-dev) by Moonshot AI, released Jun 17 2025 — 72B params, open weight.
- [o3-pro](https://aireleasetracker.com/model/openai/o3-pro) by OpenAI, released Jun 10 2025.
- [Magistral Small](https://aireleasetracker.com/model/mistral/magistral-small) by Mistral, released Jun 10 2025.
- [Magistral Medium](https://aireleasetracker.com/model/mistral/magistral-medium) by Mistral, released Jun 10 2025.
- [DeepSeek R1-0528](https://aireleasetracker.com/model/deepseek/deepseek-r1-0528) by DeepSeek, released May 28 2025 — open weight.
- [Claude Sonnet 4](https://aireleasetracker.com/model/anthropic/claude-sonnet-4) by Anthropic, released May 22 2025 — GPQA 75.4%, SWE-Bench 72.7%.
- [Claude Opus 4](https://aireleasetracker.com/model/anthropic/claude-opus-4) by Anthropic, released May 22 2025 — GPQA 79.6%, SWE-Bench 72.5%.
- [Devstral Small](https://aireleasetracker.com/model/mistral/devstral-small) by Mistral, released May 21 2025 — 24B params, open weight.
- [Mistral Medium 3](https://aireleasetracker.com/model/mistral/mistral-medium-3) by Mistral, released May 7 2025.
- [Qwen3](https://aireleasetracker.com/model/qwen/qwen3) by Qwen, released Apr 29 2025 — 235B params, 128k context, open weight.
- [Gemini 2.5 Flash](https://aireleasetracker.com/model/google/gemini-2.5-flash) by Google, released Apr 17 2025.
- [o3](https://aireleasetracker.com/model/openai/o3) by OpenAI, released Apr 16 2025.
- [o4-mini](https://aireleasetracker.com/model/openai/o4-mini) by OpenAI, released Apr 16 2025.
- [o4-mini-high](https://aireleasetracker.com/model/openai/o4-mini-high) by OpenAI, released Apr 16 2025.
- [GPT-4.1](https://aireleasetracker.com/model/openai/gpt-4.1) by OpenAI, released Apr 14 2025.
- [GPT-4.1 mini](https://aireleasetracker.com/model/openai/gpt-4.1-mini) by OpenAI, released Apr 14 2025.
- [GPT-4.1 nano](https://aireleasetracker.com/model/openai/gpt-4.1-nano) by OpenAI, released Apr 14 2025 — 1M context.
- [Kimi-VL](https://aireleasetracker.com/model/moonshot/kimi-vl) by Moonshot AI, released Apr 10 2025 — 16B params, open weight.
- [LLaMA 4 Scout](https://aireleasetracker.com/model/meta/llama-4-scout) by Meta, released Apr 5 2025 — open weight.
- [LLaMA 4 Maverick](https://aireleasetracker.com/model/meta/llama-4-maverick) by Meta, released Apr 5 2025 — open weight.
- [Gemini 2.5 Pro](https://aireleasetracker.com/model/google/gemini-2.5-pro) by Google, released Mar 25 2025 — GPQA 86.4%, SWE-Bench 59.6%, MMMU 68%.
- [DeepSeek V3-0324](https://aireleasetracker.com/model/deepseek/deepseek-v3-0324) by DeepSeek, released Mar 24 2025 — open weight.
- [o1-pro](https://aireleasetracker.com/model/openai/o1-pro) by OpenAI, released Mar 19 2025.
- [Mistral Small 3.1](https://aireleasetracker.com/model/mistral/mistral-small-3.1) by Mistral, released Mar 17 2025 — open weight.
- [Gemma 3](https://aireleasetracker.com/model/google/gemma-3) by Google, released Mar 12 2025 — open weight.
- [QwQ-32B](https://aireleasetracker.com/model/qwen/qwq-32b) by Qwen, released Mar 6 2025 — 32B params, 128k context, open weight.
- [GPT-4.5](https://aireleasetracker.com/model/openai/gpt-4.5) by OpenAI, released Feb 27 2025.
- [Claude 3.7 Sonnet](https://aireleasetracker.com/model/anthropic/claude-3.7-sonnet) by Anthropic, released Feb 24 2025 — GPQA 68%, SWE-Bench 62.3%.
- [Grok 3](https://aireleasetracker.com/model/xai/grok-3) by SpaceXAI, released Feb 17 2025.
- [Grok 3 mini](https://aireleasetracker.com/model/xai/grok-3-mini) by SpaceXAI, released Feb 17 2025.
- [Gemini 2.0 Flash-Lite](https://aireleasetracker.com/model/google/gemini-2.0-flash-lite) by Google, released Feb 5 2025.
- [Gemini 2.0 Pro](https://aireleasetracker.com/model/google/gemini-2.0-pro) by Google, released Feb 5 2025.
- [o3-mini](https://aireleasetracker.com/model/openai/o3-mini) by OpenAI, released Jan 31 2025.
- [o3-mini-high](https://aireleasetracker.com/model/openai/o3-mini-high) by OpenAI, released Jan 31 2025.
- [Gemini 2.0 Flash](https://aireleasetracker.com/model/google/gemini-2.0-flash) by Google, released Jan 30 2025.
- [Mistral Small 3](https://aireleasetracker.com/model/mistral/mistral-small-3) by Mistral, released Jan 30 2025 — open weight.
- [Qwen2.5-Max](https://aireleasetracker.com/model/qwen/qwen2.5-max) by Qwen, released Jan 28 2025 — 32k context.
- [DeepSeek Chat (R1-based)](https://aireleasetracker.com/model/deepseek/deepseek-chat-(r1-based)) by DeepSeek, released Jan 20 2025 — open weight.
- [Kimi K1.5](https://aireleasetracker.com/model/moonshot/kimi-k1.5) by Moonshot AI, released Jan 20 2025 — 128k context.
- [Codestral 25.01](https://aireleasetracker.com/model/mistral/codestral-25.01) by Mistral, released Jan 13 2025 — 256k context.
- [DeepSeek V3 Chat](https://aireleasetracker.com/model/deepseek/deepseek-v3-chat) by DeepSeek, released Dec 26 2024 — open weight.
- [DeepSeek V3 Base](https://aireleasetracker.com/model/deepseek/deepseek-v3-base) by DeepSeek, released Dec 25 2024 — open weight.
- [DeepSeek V2.5 (Revised)](https://aireleasetracker.com/model/deepseek/deepseek-v2.5-(revised)) by DeepSeek, released Dec 10 2024 — open weight.
- [LLaMA 3.3](https://aireleasetracker.com/model/meta/llama-3.3) by Meta, released Dec 6 2024 — 70B params, open weight.
- [o1](https://aireleasetracker.com/model/openai/o1) by OpenAI, released Dec 5 2024 — GPQA 75.7%, 200k context.
- [QwQ-32B-Preview](https://aireleasetracker.com/model/qwen/qwq-32b-preview) by Qwen, released Nov 28 2024 — 32B params, 32k context, open weight.
- [DeepSeek-R1-Lite Preview](https://aireleasetracker.com/model/deepseek/deepseek-r1-lite-preview) by DeepSeek, released Nov 20 2024 — open weight.
- [Mistral Large 2 24.11](https://aireleasetracker.com/model/mistral/mistral-large-2-24.11) by Mistral, released Nov 18 2024 — open weight.
- [Pixtral Large 24.11](https://aireleasetracker.com/model/mistral/pixtral-large-24.11) by Mistral, released Nov 18 2024 — open weight.
- [Qwen2.5-Coder](https://aireleasetracker.com/model/qwen/qwen2.5-coder) by Qwen, released Nov 12 2024 — 32B params, 128k context, open weight.
- [Claude 3.5 Haiku](https://aireleasetracker.com/model/anthropic/claude-3.5-haiku) by Anthropic, released Oct 22 2024 — GPQA 41.6%, SWE-Bench 40.6%.
- [Claude 3.5 Sonnet (upgraded)](https://aireleasetracker.com/model/anthropic/claude-3.5-sonnet-(upgraded)) by Anthropic, released Oct 22 2024 — SWE-Bench 49%.
- [Ministral 8B](https://aireleasetracker.com/model/mistral/ministral-8b) by Mistral, released Oct 16 2024 — open weight.
- [Ministral 3B](https://aireleasetracker.com/model/mistral/ministral-3b) by Mistral, released Oct 16 2024 — open weight.
- [Gemini 1.5 Flash-8B](https://aireleasetracker.com/model/google/gemini-1.5-flash-8b) by Google, released Oct 3 2024.
- [LLaMA 3.2](https://aireleasetracker.com/model/meta/llama-3.2) by Meta, released Sep 25 2024 — 1B/3B params, open weight.
- [Gemini 1.5 Pro-002](https://aireleasetracker.com/model/google/gemini-1.5-pro-002) by Google, released Sep 24 2024.
- [Gemini 1.5 Flash-002](https://aireleasetracker.com/model/google/gemini-1.5-flash-002) by Google, released Sep 24 2024.
- [Qwen2.5](https://aireleasetracker.com/model/qwen/qwen2.5) by Qwen, released Sep 19 2024 — 72B params, 128k context, open weight.
- [o1-preview](https://aireleasetracker.com/model/openai/o1-preview) by OpenAI, released Sep 12 2024 — GPQA 73.3%, 128k context.
- [o1-mini](https://aireleasetracker.com/model/openai/o1-mini) by OpenAI, released Sep 12 2024 — GPQA 60%, 128k context.
- [Pixtral 24.09](https://aireleasetracker.com/model/mistral/pixtral-24.09) by Mistral, released Sep 11 2024 — open weight.
- [DeepSeek V2.5](https://aireleasetracker.com/model/deepseek/deepseek-v2.5) by DeepSeek, released Sep 5 2024 — open weight.
- [GLM-4-Plus](https://aireleasetracker.com/model/zai/glm-4-plus) by Z.ai, released Aug 29 2024.
- [Grok‑2](https://aireleasetracker.com/model/xai/grok‑2) by SpaceXAI, released Aug 14 2024.
- [Mistral Large 2](https://aireleasetracker.com/model/mistral/mistral-large-2) by Mistral, released Jul 24 2024 — open weight.
- [LLaMA 3.1](https://aireleasetracker.com/model/meta/llama-3.1) by Meta, released Jul 23 2024 — open weight.
- [GPT-4o mini](https://aireleasetracker.com/model/openai/gpt-4o-mini) by OpenAI, released Jul 18 2024 — GPQA 40.2%, 128k context.
- [Mistral NeMo](https://aireleasetracker.com/model/mistral/mistral-nemo) by Mistral, released Jul 18 2024 — 12B params, 128k context, open weight.
- [Codestral Mamba 7B](https://aireleasetracker.com/model/mistral/codestral-mamba-7b) by Mistral, released Jul 16 2024 — open weight.
- [Mathstral 7B](https://aireleasetracker.com/model/mistral/mathstral-7b) by Mistral, released Jul 16 2024 — open weight.
- [Gemma 2](https://aireleasetracker.com/model/google/gemma-2) by Google, released Jun 27 2024 — open weight.
- [Claude 3.5 Sonnet](https://aireleasetracker.com/model/anthropic/claude-3.5-sonnet) by Anthropic, released Jun 20 2024 — GPQA 59.4%, SWE-Bench 33.4%.
- [DeepSeek Coder V2](https://aireleasetracker.com/model/deepseek/deepseek-coder-v2) by DeepSeek, released Jun 17 2024 — open weight.
- [Qwen2](https://aireleasetracker.com/model/qwen/qwen2) by Qwen, released Jun 6 2024 — 72B params, 128k context, open weight.
- [GLM-4-9B](https://aireleasetracker.com/model/zai/glm-4-9b) by Z.ai, released Jun 5 2024 — 9B params, open weight.
- [Codestral 22B](https://aireleasetracker.com/model/mistral/codestral-22b) by Mistral, released May 29 2024.
- [Gemini 1.5 Flash](https://aireleasetracker.com/model/google/gemini-1.5-flash) by Google, released May 14 2024.
- [GPT-4o](https://aireleasetracker.com/model/openai/gpt-4o) by OpenAI, released May 13 2024 — GPQA 49.9%, 128k context.
- [DeepSeek V2](https://aireleasetracker.com/model/deepseek/deepseek-v2) by DeepSeek, released May 6 2024 — open weight.
- [LLaMA 3 (8B/70B)](https://aireleasetracker.com/model/meta/llama-3-(8b/70b)) by Meta, released Apr 18 2024 — open weight.
- [Mixtral 8×22B](https://aireleasetracker.com/model/mistral/mixtral-8×22b) by Mistral, released Apr 10 2024 — open weight.
- [Grok‑1.5](https://aireleasetracker.com/model/xai/grok‑1.5) by SpaceXAI, released Mar 29 2024.
- [Claude 3 Haiku](https://aireleasetracker.com/model/anthropic/claude-3-haiku) by Anthropic, released Mar 4 2024 — GPQA 33.3%.
- [Claude 3 Sonnet](https://aireleasetracker.com/model/anthropic/claude-3-sonnet) by Anthropic, released Mar 4 2024 — GPQA 40.4%.
- [Claude 3 Opus](https://aireleasetracker.com/model/anthropic/claude-3-opus) by Anthropic, released Mar 4 2024 — GPQA 50.4%, SWE-Bench 33%.
- [Mistral Large](https://aireleasetracker.com/model/mistral/mistral-large) by Mistral, released Feb 26 2024.
- [Mistral Small](https://aireleasetracker.com/model/mistral/mistral-small) by Mistral, released Feb 26 2024.
- [Gemma 1](https://aireleasetracker.com/model/google/gemma-1) by Google, released Feb 21 2024 — open weight.
- [Gemini 1.5 Pro](https://aireleasetracker.com/model/google/gemini-1.5-pro) by Google, released Feb 15 2024.
- [Gemini 1.0 Ultra](https://aireleasetracker.com/model/google/gemini-1.0-ultra) by Google, released Feb 8 2024.
- [DeepSeek-Math](https://aireleasetracker.com/model/deepseek/deepseek-math) by DeepSeek, released Feb 5 2024 — open weight.
- [Qwen1.5](https://aireleasetracker.com/model/qwen/qwen1.5) by Qwen, released Feb 4 2024 — 72B params, 32k context, open weight.
- [Code Llama 70B](https://aireleasetracker.com/model/meta/code-llama-70b) by Meta, released Jan 29 2024 — open weight.
- [GLM-4](https://aireleasetracker.com/model/zai/glm-4) by Z.ai, released Jan 16 2024.
- [DeepSeek-MoE](https://aireleasetracker.com/model/deepseek/deepseek-moe) by DeepSeek, released Jan 11 2024 — open weight.
- [Gemini 1.0 Pro](https://aireleasetracker.com/model/google/gemini-1.0-pro) by Google, released Dec 13 2023.
- [Mistral Medium](https://aireleasetracker.com/model/mistral/mistral-medium) by Mistral, released Dec 11 2023.
- [Mixtral 8×7B](https://aireleasetracker.com/model/mistral/mixtral-8×7b) by Mistral, released Dec 8 2023 — open weight.
- [Gemini 1.0 Nano](https://aireleasetracker.com/model/google/gemini-1.0-nano) by Google, released Dec 6 2023.
- [Qwen-72B](https://aireleasetracker.com/model/qwen/qwen-72b) by Qwen, released Nov 30 2023 — 72B params, 32k context, open weight.
- [DeepSeek-LLM](https://aireleasetracker.com/model/deepseek/deepseek-llm) by DeepSeek, released Nov 29 2023 — open weight.
- [Claude 2.1](https://aireleasetracker.com/model/anthropic/claude-2.1) by Anthropic, released Nov 21 2023 — 200k context.
- [GPT-4 Turbo](https://aireleasetracker.com/model/openai/gpt-4-turbo) by OpenAI, released Nov 6 2023 — GPQA 42.5%, 128k context.
- [Grok‑1](https://aireleasetracker.com/model/xai/grok‑1) by SpaceXAI, released Nov 3 2023 — open weight.
- [DeepSeek Coder](https://aireleasetracker.com/model/deepseek/deepseek-coder) by DeepSeek, released Nov 2 2023 — open weight.
- [Kimi Chat](https://aireleasetracker.com/model/moonshot/kimi-chat) by Moonshot AI, released Oct 9 2023.
- [Mistral 7B](https://aireleasetracker.com/model/mistral/mistral-7b) by Mistral, released Sep 27 2023 — open weight.
- [Code Llama](https://aireleasetracker.com/model/meta/code-llama) by Meta, released Aug 24 2023 — open weight.
- [Claude Instant 1.2](https://aireleasetracker.com/model/anthropic/claude-instant-1.2) by Anthropic, released Aug 9 2023 — 100k context.
- [Qwen-7B](https://aireleasetracker.com/model/qwen/qwen-7b) by Qwen, released Aug 3 2023 — 7B params, 2k context, open weight.
- [LLaMA 2](https://aireleasetracker.com/model/meta/llama-2) by Meta, released Jul 18 2023 — open weight.
- [Claude 2](https://aireleasetracker.com/model/anthropic/claude-2) by Anthropic, released Jul 11 2023 — 100k context.
- [Bard](https://aireleasetracker.com/model/google/bard) by Google, released Mar 21 2023.
- [Claude 1](https://aireleasetracker.com/model/anthropic/claude-1) by Anthropic, released Mar 14 2023 — 9k context.
- [GPT-4](https://aireleasetracker.com/model/openai/gpt-4) by OpenAI, released Mar 14 2023 — 8k context.
- [ChatGLM-6B](https://aireleasetracker.com/model/zai/chatglm-6b) by Z.ai, released Mar 14 2023 — 6B params, open weight.
- [LLaMA 1](https://aireleasetracker.com/model/meta/llama-1) by Meta, released Feb 24 2023 — open weight.
- [GPT-3.5](https://aireleasetracker.com/model/openai/gpt-3.5) by OpenAI, released Nov 30 2022.

## Citation

When citing this site, please link to https://aireleasetracker.com and the specific model or company page. All release dates and benchmark scores are sourced from official announcements by the releasing organizations.