Frontier Log

Models

OpenAI

ModelReleasedResultsBenchmarks
GPT-6.1 Sol29 Sep 2026None
GPT-6 Sol and Luna22 Sep 2026None
GPT-6 Astra3 Sep 202612HLE, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 4.0, TB-Science 0.1, AutomationBench, OSWorld 2.0, FrontierCode 1.1, DeepSWE v1.1, ARC-AGI-1, ARC-AGI-3
GPT-5.6-Cyber10 Aug 2026None
GPT-5.6 Luna for free users6 Aug 2026None
GPT-5.6 Sol9 Jul 20269SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 4.0, OSWorld 2.0, DeepSWE v1.1, Terminal-Bench 2.1, ARC-AGI-3
GPT-Live8 Jul 2026None
GPT-5.6 Sol preview26 Jun 2026None
Realtime voice models in the API7 May 2026None
GPT-5.5-Cyber7 May 2026None
GPT-5.5 Instant5 May 2026None
GPT-5.523 Apr 20269HLE, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, OSWorld-Verified, ARC-AGI-1
OpenAI Privacy Filter22 Apr 2026None
GPT-Rosalind16 Apr 2026None
GPT-5.4-Cyber for vetted defenders14 Apr 2026None
GPT-5.4 mini and nano17 Mar 2026None
GPT-5.45 Mar 20269HLE, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, OSWorld-Verified, ARC-AGI-1
GPT-5.3 Instant3 Mar 2026None
GPT-5.3-Codex-Spark12 Feb 2026None
GPT-5.3-Codex5 Feb 20261SWE-bench Pro
GPT-5.2-Codex18 Dec 2025None
ChatGPT Images and GPT-Image-1.516 Dec 2025None
GPT-5.211 Dec 202511HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, AIME 2025, ARC-AGI-1, HMMT Feb 2025
GPT-5.1-Codex-Max19 Nov 2025None
GPT-5.1 in the API13 Nov 2025None
GPT-5.112 Nov 20255HLE, SWE-bench Verified, GPQA Diamond, ARC-AGI-2
gpt-oss-safeguard29 Oct 2025None
Sora 230 Sep 2025None
gpt-realtime28 Aug 2025None
GPT-57 Aug 20255SWE-bench Verified, AIME 2025
gpt-oss-120b5 Aug 20251GPQA Diamond
OpenAI o316 Apr 20255SWE-bench Verified, AIME 2024, AIME 2025
GPT-4.114 Apr 2025None
Next-generation audio models in the API20 Mar 2025None
GPT-4.527 Feb 20253SWE-bench Verified, GPQA Diamond, AIME 2024
OpenAI o3-mini31 Jan 2025None
o1 for developers17 Dec 2024None
o1-mini12 Sep 2024None
OpenAI o112 Sep 2024None
GPT-4o mini18 Jul 2024None
GPT-4o13 May 20242SWE-bench Verified, GPQA Diamond
Sora announced15 Feb 2024None
GPT-4 Turbo6 Nov 2023None
GPT-414 Mar 2023None
Whisper21 Sep 2022None
InstructGPT27 Jan 2022None
Codex10 Aug 2021None
DALL·E5 Jan 2021None
CLIP5 Jan 2021None
GPT-328 May 2020None
GPT-214 Feb 2019None
GPT-111 Jun 2018None

Anthropic

ModelReleasedResultsBenchmarks
Claude Haiku 5.57 Oct 20263HLE, SWE-bench Pro, Terminal-Bench 4.0
Claude Sonnet 5.528 Sep 20264HLE, SWE-bench Pro, Terminal-Bench 4.0, DeepSWE v1.1
Claude Opus 5.522 Sep 20268HLE, SWE-bench Pro, Terminal-Bench 4.0, TB-Science 0.1, AutomationBench, FrontierCode 1.1, DeepSWE v1.1
Claude Fable 5.11 Sep 20268HLE, SWE-bench Pro, ARC-AGI-2, Terminal-Bench 4.0, OSWorld 2.0, DeepSWE v1.1
Claude Opus 524 Jul 202611HLE, SWE-bench Pro, BrowseComp, ARC-AGI-2, Terminal-Bench 4.0, OSWorld 2.0, DeepSWE v1.1, ARC-AGI-3
Claude Sonnet 530 Jun 20264HLE, SWE-bench Pro, BrowseComp
Claude Mythos 59 Jun 20266HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp
Claude Fable 59 Jun 20268HLE, SWE-bench Verified, SWE-bench Pro, BrowseComp, DeepSWE v1.1
Claude Opus 4.828 May 20269HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, DeepSWE v1.1, Terminal-Bench 2.1, OSWorld-Verified
Claude Opus 4.716 Apr 202610HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, OSWorld-Verified
Claude Mythos Preview7 Apr 20267HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp
Claude Sonnet 4.617 Feb 2026None
Claude Opus 4.65 Feb 202611HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, τ²-bench
Claude Opus 4.524 Nov 20258HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0
Claude Haiku 4.515 Oct 20251HLE
Claude Sonnet 4.529 Sep 20259HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0
Claude Opus 4.15 Aug 20256HLE, SWE-bench Verified, GPQA Diamond, AIME 2025, Terminal-Bench 1.0
Claude Gov models6 Jun 2025None
Claude Sonnet 422 May 20255SWE-bench Verified, GPQA Diamond
Claude Opus 422 May 202511SWE-bench Verified, GPQA Diamond, AIME 2025, Terminal-Bench 1.0
Claude 3.7 Sonnet24 Feb 20258SWE-bench Verified, GPQA Diamond, AIME 2024
Upgraded Claude 3.5 Sonnet and 3.5 Haiku22 Oct 2024None
Claude 3.5 Sonnet21 Jun 20242GPQA Diamond
Claude 3 Haiku13 Mar 2024None
Claude 3 Sonnet4 Mar 20241GPQA Diamond
Claude 3 Opus4 Mar 20241GPQA Diamond
Claude 3 Haiku4 Mar 20241GPQA Diamond
Claude 2.121 Nov 2023None
Claude Instant 1.29 Aug 2023None
Claude 211 Jul 2023None
Claude and Claude Instant14 Mar 2023None

Google

ModelReleasedResultsBenchmarks
Gemini 4 Argon generally available13 Oct 2026None
Gemini 4 Argon30 Sep 20265Terminal-Bench 4.0, TB-Science 0.1, AutomationBench, OSWorld 2.0, DeepSWE v1.1
Gemini 3.8 Live15 Sep 2026None
WeatherNext 33 Sep 2026None
Gemini 3.8 Flash2 Sep 20261DeepSWE v1.1
Gemini Omni 1.1 Flash27 Aug 2026None
Gemini 3.7 Flash13 Aug 20261DeepSWE v1.1
Gemini Robotics 230 Jul 2026None
Lyria 3.529 Jul 2026None
Gemini 3.6 Flash21 Jul 20263SWE-bench Pro, DeepSWE v1.1
Gemini 3.5 Flash Cyber21 Jul 2026None
DiffusionGemma10 Jun 2026None
Gemma 4 12B3 Jun 2026None
Gemini Omni Flash19 May 2026None
Gemini 3.5 Flash19 May 20264HLE, SWE-bench Pro, ARC-AGI-2, DeepSWE v1.1
Gemini Robotics-ER 1.614 Apr 2026None
Gemma 4 31B2 Apr 20261GPQA Diamond
Gemini 3.1 Flash Live26 Mar 2026None
Gemini Embedding 210 Mar 2026None
Gemini 3.1 Flash-Lite3 Mar 2026None
Nano Banana 226 Feb 2026None
Gemini 3.1 Pro19 Feb 20269HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0, DeepSWE v1.1
Lyria 318 Feb 2026None
Gemini 3 Deep Think upgrade12 Feb 2026None
TranslateGemma15 Jan 2026None
MedGemma 1.5 and MedASR13 Jan 2026None
T5Gemma 218 Dec 2025None
FunctionGemma18 Dec 2025None
Gemini 3 Flash17 Dec 2025None
Nano Banana Pro20 Nov 2025None
Gemini 3 Pro18 Nov 20259HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, ARC-AGI-2, Terminal-Bench 2.0
WeatherNext 217 Nov 2025None
SIMA 213 Nov 2025None
Veo 3.115 Oct 2025None
C2S-Scale 27B15 Oct 2025None
Gemini 2.5 Computer Use model7 Oct 2025None
Gemini Robotics 1.5 and ER 1.525 Sep 2025None
VaultGemma12 Sep 2025None
Nano Banana26 Aug 2025None
Gemma 3 270M14 Aug 2025None
Genie 35 Aug 2025None
AlphaEarth Foundations30 Jul 2025None
T5Gemma9 Jul 2025None
MedGemma 27B Multimodal and MedSigLIP9 Jul 2025None
Gemma 3n26 Jun 2025None
AlphaGenome25 Jun 2025None
Gemini Robotics On-Device24 Jun 2025None
Gemini 2.5 GA and 2.5 Flash-Lite17 Jun 2025None
Veo 3 and Imagen 420 May 2025None
Gemini Diffusion20 May 2025None
AlphaEvolve14 May 2025None
Gemini 2.5 Flash17 Apr 2025None
TxGemma25 Mar 2025None
Gemini 2.5 Pro25 Mar 20259HLE, SWE-bench Verified, GPQA Diamond, ARC-AGI-2, AIME 2024, AIME 2025
Gemma 3 27B12 Mar 20251GPQA Diamond
Gemini Robotics and Gemini Robotics-ER12 Mar 2025None
Gemini 2.0 Pro5 Feb 20251GPQA Diamond
Gemini 2.0 Flash-Lite5 Feb 20251GPQA Diamond
Veo 216 Dec 2024None
Gemini 2.0 Flash11 Dec 20242GPQA Diamond
Genie 24 Dec 2024None
GenCast4 Dec 2024None
AlphaChip26 Sep 2024None
DataGemma12 Sep 2024None
AlphaProteo5 Sep 2024None
Gemma 227 Jun 2024None
Veo14 May 2024None
Imagen 314 May 2024None
Gemini 1.5 Flash14 May 2024None
AlphaFold 38 May 2024None
Gemma21 Feb 2024None
Gemini 1.515 Feb 2024None
Imagen 21 Feb 2024None
AlphaGeometry17 Jan 2024None
Gemini 1.0 (Ultra, Pro and Nano)6 Dec 2023None
GNoME materials discovery29 Nov 2023None
Lyria16 Nov 2023None
GraphCast14 Nov 2023None
AlphaMissense19 Sep 2023None
RT-228 Jul 2023None
PaLM 210 May 2023None
RT-1 Robotics Transformer13 Dec 2022None
AlphaCode8 Dec 2022None
Gato: a generalist agent12 May 2022None
PaLM4 Apr 2022None
GLaM9 Dec 2021None
LaMDA18 May 2021None
Vision Transformer (ViT)3 Dec 2020None
AlphaFold 2: protein folding solved30 Nov 2020None
T5: text-to-text Transformer24 Feb 2020None
AlphaZero6 Dec 2018None
BERT2 Nov 2018None
AlphaGo Zero18 Oct 2017None
Google Neural Machine Translation27 Sep 2016None
WaveNet8 Sep 2016None
AlphaGo27 Jan 2016None

Meta

ModelReleasedResultsBenchmarks
Muse Spark Open15 Oct 2026None
Muse Wonder and Muse Wonder Full12 Oct 2026None
Muse Spark 1.32 Sep 2026None
Muse Glimmer12 Aug 2026None
Muse Spark 1.2 and Muse Code5 Aug 2026None
Muse Spark 1.19 Jul 2026None
Muse Spark8 Apr 2026None
Llama 4 Scout and Maverick5 Apr 2025None
Llama 3.1 405B23 Jul 2024None
Llama 218 Jul 2023None
LLaMA24 Feb 2023None

DeepSeek

ModelReleasedResultsBenchmarks
DeepSeek-V4.1-Flash-101717 Oct 20262Terminal-Bench 4.0, DeepSWE v1.1
DeepSeek-V4.1-Flash10 Sep 20263Terminal-Bench 4.0, DeepSWE v1.1, Terminal-Bench 2.1
DeepSeek-V4-Flash-Vision-Exp21 Aug 2026None
DeepSeek-V4-Pro GA13 Aug 2026None
DeepSeek-V4-Pro24 Apr 20267SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, MMLU-Pro, LiveCodeBench
DeepSeek-V3.21 Dec 20254HLE, SWE-bench Verified, Terminal-Bench 2.0, AIME 2025
DeepSeek-V3.2-Exp29 Sep 20254SWE-bench Verified, BrowseComp, AIME 2025, Terminal-Bench 1.0
DeepSeek-V3.1-Terminus22 Sep 2025None
DeepSeek-V3.121 Aug 20257HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, AIME 2025, Terminal-Bench 1.0, LiveCodeBench
DeepSeek-R1-052828 May 2025None
DeepSeek-V3-032425 Mar 2025None
DeepSeek-R120 Jan 20252SWE-bench Verified, AIME 2024
DeepSeek-V326 Dec 20241GPQA Diamond

Mistral

ModelReleasedResultsBenchmarks
Mistral Large 4 (public preview)6 Oct 2026None
Leanstral 1.52 Jul 2026None
Mistral Medium 3.522 May 20261SWE-bench Verified
Mistral Small 416 Mar 2026None
Leanstral16 Mar 2026None
Devstral 29 Dec 20251SWE-bench Verified
Mistral 32 Dec 2025None
Codestral 25.0830 Jul 2025None
Voxtral15 Jul 2025None
Devstral Small 1.110 Jul 2025None
Magistral10 Jun 2025None
Devstral21 May 2025None
Mistral Medium 37 May 2025None
Mistral Small 3.117 Mar 2025None
Mistral Saba17 Feb 2025None
Mistral Small 330 Jan 2025None
Codestral 25.0113 Jan 2025None
Mixtral 8x7B11 Dec 2023None

Alibaba

ModelReleasedResultsBenchmarks
Qwen3.8-Omni-Flash18 Sep 2026None
Qwen3.8-Flash-Next26 Aug 2026None
Qwen3.8-Max3 Aug 20267HLE, SWE-bench Pro, GPQA Diamond, OSWorld 2.0, DeepSWE v1.1, Terminal-Bench 2.1, OSWorld-Verified
Qwen-AgentWorld23 Jun 2026None
Qwen3.7-Plus1 Jun 2026None
Qwen3.7-Max20 May 202612HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, Terminal-Bench 2.0, AutomationBench, DeepSWE v1.1, Terminal-Bench 2.1, MMLU-Pro, LiveCodeBench, MCP Atlas
Qwen3.6-Max-Preview18 Apr 20262SWE-bench Pro, Terminal-Bench 2.0
Qwen3.6-35B-A3B15 Apr 2026None
Qwen3.6-Plus2 Apr 202612HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, Terminal-Bench 2.0, OSWorld-Verified, MMLU-Pro, HMMT Feb 2025, LiveCodeBench, MCP Atlas
Qwen3.5-Omni30 Mar 2026None
Qwen3.5-397B-A17B16 Feb 202614HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, OSWorld-Verified, MMLU-Pro, HMMT Feb 2025, LiveCodeBench, MCP Atlas
Qwen3-Coder-Next3 Feb 2026None
Qwen3-Max-Thinking26 Jan 202613HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, MMLU-Pro, HMMT Feb 2025, LiveCodeBench
Qwen3-Max24 Sep 20253SWE-bench Verified, AIME 2025, LiveCodeBench
Qwen3-VL23 Sep 2025None
Qwen3-Omni22 Sep 2025None
Qwen3-Next11 Sep 2025None
Qwen3-235B-A22B-Thinking-250725 Jul 2025None
Qwen3-Coder22 Jul 2025None
Qwen3-235B-A22B29 Apr 20253AIME 2024, AIME 2025, LiveCodeBench
QVQ-Max28 Mar 2025None
Qwen2.5-Omni27 Mar 2025None
QwQ-32B6 Mar 20252AIME 2024, LiveCodeBench
QwQ-Max-Preview25 Feb 2025None
Qwen2.5-Max28 Jan 20253GPQA Diamond, MMLU-Pro, LiveCodeBench
Qwen2.5-1M27 Jan 2025None
Qwen2.5-VL26 Jan 2025None

Moonshot AI

ModelReleasedResultsBenchmarks
Kimi K316 Jul 20266HLE, BrowseComp, DeepSWE v1.1, MCP Atlas
Kimi K2.620 Apr 20263HLE, SWE-bench Pro, BrowseComp
Kimi K2.527 Jan 20265HLE, SWE-bench Verified, SWE-bench Pro, BrowseComp
Kimi K2 Thinking6 Nov 20252HLE, BrowseComp
Kimi K211 Jul 20252SWE-bench Verified, GPQA Diamond

MiniMax

ModelReleasedResultsBenchmarks
MiniMax M31 Jun 20263SWE-bench Verified, SWE-bench Pro, BrowseComp
MiniMax M2.718 Mar 20267SWE-bench Verified, SWE-bench Pro, BrowseComp, Terminal-Bench 2.0, Terminal-Bench 2.1, MCP Atlas
MiniMax M2.512 Feb 20263SWE-bench Verified, SWE-bench Pro, BrowseComp
MiniMax M2.123 Dec 20256SWE-bench Verified, BrowseComp, Terminal-Bench 2.0
MiniMax M227 Oct 20257SWE-bench Verified, BrowseComp, Terminal-Bench 2.0, Terminal-Bench 1.0
MiniMax-M116 Jun 20257HLE, SWE-bench Verified, GPQA Diamond, AIME 2024, AIME 2025, MMLU-Pro, LiveCodeBench
MiniMax-Text-0115 Jan 20252GPQA Diamond, MMLU-Pro

Z.ai

ModelReleasedResultsBenchmarks
GLM-5.3-Flash26 Aug 2026None
GLM-5.314 Aug 20264HLE, AutomationBench, DeepSWE v1.1, Terminal-Bench 2.1
GLM-5.216 Jun 202610HLE, SWE-bench Pro, GPQA Diamond, AutomationBench, DeepSWE v1.1, Terminal-Bench 2.1, MCP Atlas
GLM-5.17 Apr 202612HLE, SWE-bench Pro, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, DeepSWE v1.1, Terminal-Bench 2.1, MCP Atlas
GLM-512 Feb 202611HLE, SWE-bench Verified, SWE-bench Pro, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, MCP Atlas
GLM-4.722 Dec 202513HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, AIME 2025, MMLU-Pro, HMMT Feb 2025, LiveCodeBench, MCP Atlas
GLM-4.6V8 Dec 2025None
GLM-4.630 Sep 202513HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, Terminal-Bench 2.0, AIME 2025, Terminal-Bench 1.0, MMLU-Pro, HMMT Feb 2025, LiveCodeBench
GLM-4.528 Jul 202511HLE, SWE-bench Verified, GPQA Diamond, BrowseComp, AIME 2024, AIME 2025, Terminal-Bench 1.0, MMLU-Pro, LiveCodeBench

NVIDIA

ModelReleasedResultsBenchmarks
Nemotron 3.5 Super VL and Bolt15 Oct 2026None
Nemotron 3.5 Lightning11 Aug 2026None
Nemotron 3 Ultra4 Jun 2026None
Nemotron 3 Nano Omni28 Apr 2026None
Nemotron 3 Super11 Mar 2026None
Nemotron 315 Dec 2025None
Nemotron Nano 2 VL28 Oct 2025None
Nemotron Nano 218 Aug 2025None
Llama Nemotron Super v1.529 Jul 2025None
Llama Nemotron18 Mar 2025None

xAI

ModelReleasedResultsBenchmarks
Grok-117 Mar 2024None

Snowflake

ModelReleasedResultsBenchmarks
Snowflake Arctic24 Apr 2024None

Sber

ModelReleasedResultsBenchmarks
GigaChat 3 Ultra Preview21 Nov 2025None

Thinking Machines

ModelReleasedResultsBenchmarks
Inkling15 Jul 20262SWE-bench Verified, SWE-bench Pro

Poolside

ModelReleasedResultsBenchmarks
Laguna XS 2.12 Jul 20262SWE-bench Verified, SWE-bench Pro