Analytics

How the field moves, built only from measured results and dated releases. See methodology.

Best Intelligence Index over time

Running best score among models released so far, open-weight versus proprietary. 57 models with a measured index and a trustworthy release date; Artificial Analysis mostly indexes models released from mid-2026, so earlier history is thin.

Show the data as a table
Release dateModelLabOpen weightsIndex
2025-08-05gpt-oss-120bOpenAIYes3.8
2026-03-16Mistral Small 4Mistral AIYes4.3
2026-03-16NVIDIA Nemotron 3 Nano 4BNVIDIA AI & NIMYes1.8
2026-04-02Gemma 4 31BGoogle DeepMindYes4.4
2026-04-29Mistral Medium 3.5Mistral AIYes4.6
2026-06-01MiniMax M3MiniMaxYes18.9
2026-06-04Nemotron 3 UltraNVIDIA AI & NIMYes11.8
2026-06-09North Mini CodeCohereYes3.4
2026-07-15InklingThinking Machines LabYes12.5
2026-07-21Gemini 3.5 Flash-LiteGoogle DeepMindNo11.7
2026-07-30Inkling SmallThinking Machines LabYes12.9
2026-08-06Solar Pro 4UpstageNo14.2
2026-08-10Muse GlimmerMeta AIYes6.3
2026-08-11Nemotron 3.5 LightningNVIDIA AI & NIMYes5.5
2026-08-12K-EXAONE 2.0LG AI ResearchYes7.6
2026-08-13DeepSeek V4-ProDeepSeekYes23.1
2026-08-14Qwen3.8 27BAlibaba Cloud / QwenYes20.1
2026-08-18GLM 5.3Zhipu AI (Z.ai)Yes31.7
2026-08-25Granite 4.2 8BIBMYes5.1
2026-08-25Granite 4.2 3BIBMYes4.7
2026-08-26GLM 5.3 FlashZhipu AI (Z.ai)Yes27.8
2026-08-26Qwen3.8-Flash-NextAlibaba Cloud / QwenYes28.6
2026-08-30Apodex 1.1ApodexNo13.7
2026-09-02Qwen3.8 Max (0902)Alibaba Cloud / QwenYes33.7
2026-09-02Gemini 3.8 FlashGoogle DeepMindNo30.8
2026-09-03GPT-6 AstraOpenAINo56.9
2026-09-03K2 Horizon 375B A23BInstitute of Foundation Models (MBZUAI)Yes17.8
2026-09-18Step 5StepFunYes28.5
2026-09-21Grok 4.7xAINo35.5
2026-09-21MiMo-V2.6-ProXiaomiYes29.9
2026-09-21MiMo-V2.6-FlashXiaomiYes26.7
2026-09-22Claude Opus 5.5AnthropicNo59.1
2026-09-22GPT-6 LunaOpenAINo27.9
2026-09-28Claude Sonnet 5.5AnthropicNo54.5
2026-09-29GPT-6.1 SolOpenAINo54.6
2026-09-30Gemini 4 ArgonGoogle DeepMindNo50.4
2026-10-07Claude Haiku 5.5AnthropicNo36.5
2026-10-12Muse Wonder FullMeta AINo50.1
2026-10-12Muse WonderMeta AINo43.6
2026-10-15Nemotron 3.5 Super VLNVIDIA AI & NIMYes13.6
2026-10-15Muse Spark OpenMeta AIYes24.9
2026-10-16Beam 1.0Reflection AIYes17.7
2026-10-17DeepSeek V4.1 Flash 1017DeepSeekYes32.6
2026-10-19MiniMax M3.1 FlashMiniMaxYes29.1
2026-10-22Ling 3.1 Flash 1019inclusionAI (Ant Group)Yes29.6
2026-10-23Kimi K3.1 ProMoonshot AIYes46.5
2026-10-25Nanbeige4.3-33BNanbeigeYes20.4
2026-10-26GPT-6.1 TerraOpenAINo51.3
2026-10-28Gemini 4 CarbonGoogle DeepMindNo60.0
2026-10-29Mistral Large 4Mistral AIYes26.1
2026-11-03Hy4TencentNo34.6
2026-11-05Inkling 2 SmallThinking Machines LabYes30.6
2026-11-07Agnes 2.5 Pro ThetaSapiens AINo24.0
2026-11-09Claude Fable 5.2AnthropicNo64.6
2026-11-09Muse EmberMeta AIYes36.0
2026-11-11JT-4.1 Flash 236B A21BChina MobileNo19.0
2026-11-12Gemini 4 HydrogenGoogle DeepMindNo53.9

Source: Artificial Analysis Intelligence Index; release dates from the Artificial Analysis tracker and verified announcements

64.6best proprietary indexClaude Fable 5.2
46.5best open-weight indexKimi K3.1 Pro
18.1 pointsgap between them
57models in the time series2025-08-05 to 2026-11-12

Price against performance

Intelligence Index against what it cost Artificial Analysis to run the whole Index, for 48 models. Up and to the left is better. Cheapest model within 15% of the top score: Gemini 4 Carbon at $5.48.

Show the data as a table
ModelLabOpen weightsIndexCost to run Index (USD)
Claude Fable 5.2AnthropicNo64.626.54
Gemini 4 CarbonGoogle DeepMindNo60.05.48
Claude Opus 5.5AnthropicNo59.110.04
GPT-6 AstraOpenAINo56.96.24
GPT-6.1 SolOpenAINo54.61.27
Claude Sonnet 5.5AnthropicNo54.510.02
Gemini 4 HydrogenGoogle DeepMindNo53.93.16
GPT-6.1 TerraOpenAINo51.30.96
Gemini 4 ArgonGoogle DeepMindNo50.44.97
Kimi K3.1 ProMoonshot AIYes46.53.68
Muse WonderMeta AINo43.66.95
Claude Haiku 5.5AnthropicNo36.50.41
Muse EmberMeta AIYes36.00.14
Grok 4.7xAINo35.57.77
Hy4TencentNo34.61.03
Qwen3.8 Max (0902)Alibaba Cloud / QwenYes33.710.02
DeepSeek V4.1 Flash 1017DeepSeekYes32.60.60
GLM 5.3Zhipu AI (Z.ai)Yes31.75.01
Gemini 3.8 FlashGoogle DeepMindNo30.87.77
Inkling 2 SmallThinking Machines LabYes30.61.18
MiMo-V2.6-ProXiaomiYes29.90.26
Ling 3.1 Flash 1019inclusionAI (Ant Group)Yes29.63.19
MiniMax M3.1 FlashMiniMaxYes29.10.49
Qwen3.8-Flash-NextAlibaba Cloud / QwenYes28.60.57
Step 5StepFunYes28.52.35
GPT-6 LunaOpenAINo27.90.15
GLM 5.3 FlashZhipu AI (Z.ai)Yes27.80.58
MiMo-V2.6-FlashXiaomiYes26.70.13
Mistral Large 4Mistral AIYes26.12.06
Muse Spark OpenMeta AIYes24.92.02
Agnes 2.5 Pro ThetaSapiens AINo24.00.57
DeepSeek V4-ProDeepSeekYes23.11.77
Qwen3.8 27BAlibaba Cloud / QwenYes20.13.53
MiniMax M3MiniMaxYes18.91.29
Beam 1.0Reflection AIYes17.70.28
Solar Pro 4UpstageNo14.20.96
Apodex 1.1ApodexNo13.70.76
Nemotron 3.5 Super VLNVIDIA AI & NIMYes13.60.56
Inkling SmallThinking Machines LabYes12.90.20
InklingThinking Machines LabYes12.51.35
Nemotron 3 UltraNVIDIA AI & NIMYes11.82.21
Gemini 3.5 Flash-LiteGoogle DeepMindNo11.70.69
Muse GlimmerMeta AIYes6.30.18
Nemotron 3.5 LightningNVIDIA AI & NIMYes5.50.31
Granite 4.2 8BIBMYes5.10.04
Mistral Medium 3.5Mistral AIYes4.61.04
Mistral Small 4Mistral AIYes4.30.04
gpt-oss-120bOpenAIYes3.80.35

Source: Artificial Analysis (Index cost per run)

Model releases per quarter

Releases with a verified, Artificial Analysis or version-id date, over the last 8 quarters in the data (351 models). Eight labs have a fixed colour; everything else is grouped. Models whose only date is an OpenRouter listing are not counted.

Show the data as a table
QuarterGoogle DeepMindAnthropicDeepSeekAlibaba Cloud / QwenMeta AIOpenAIMoonshot AIxAIOther labs
2025 Q16113021119
2025 Q2200121003
2025 Q31204020115
2025 Q43210072031
2026 Q182013060227
2026 Q211435052226
2026 Q364697141340
2026 Q44210421017

Source: Verified announcements, Artificial Analysis release tracker, OpenRouter version ids