Artificial Analysis Intelligence Index
Composite intelligence score across AA evaluations
v4.3.2composite indexactive
Score, AA leaderboard, 205 current models
Definition
- Version
- v4.3.2
- Category
- composite index
- Maintainer
- Artificial Analysis
- Status
- active
- Home page
- https://artificialanalysis.ai/leaderboards/models
Sources
- LLM Leaderboard - Comparison of AI models from OpenAI, Anthropic, Google, SpaceXAI & others | Artificial Analysis artificialanalysis.ai/leaderboards/models https://artificialanalysis.ai/leaderboards/models
- Claude Opus 5.5 (max with fallback) - Intelligence, Performance & Price Analysis | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/claude-opus-5-5
- Best AI models, October 2026: Opus 5.5, GPT-6, Gemini slash-digital.io https://slash-digital.io/en/insights/ai-models-2026/
- LLM Leaderboard - Comparison of AI models from OpenAI, Anthropic, Google, SpaceXAI & others | Artificial Analysis artificialanalysis.ai/leaderboards/models?open_weights=true https://artificialanalysis.ai/leaderboards/models?open_weights=true
- Comparison of Open Source AI Models across Intelligence, Performance, Price, Context Window, and more | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/open-source
- AI Model & API Providers Analysis | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/
- datalearner.com/en/leaderboards/external/aa-quality-index https://www.datalearner.com/en/leaderboards/external/aa-quality-index
Scores at a Glance
Score, AA leaderboard, 205 current models
Higher is better for this unit.
1
Claude Opus 5.5Anthropicmax; 57.62 per secondary orcarouterClaude Sonnet 5.5AnthropicmaxClaude Fable 5.1Anthropicmax
Gemini 4 ArgonGoogle DeepMindhigh; 52.56 per secondary
GPT-6 AstraOpenAImax; 52.67 per secondaryGPT-6.1 SolOpenAImax
Muse Spark 1.3Metamax
Grok 4.7SpaceXAI (xAI)xhigh
MiMo-V2.6-ProXiaomi
Qwen3.8 MaxAlibaba (Qwen)0902
GLM-5.3Z.ai (Zhipu)max
Kimi K3Moonshot AImax
Step 5 PreviewStepFunClaude Haiku 5.5AnthropicmaxGemini 3.8 FlashGoogle DeepMindhigh; 40.93
DeepSeek-V4.1-FlashDeepSeekmax
Mistral Large 4 (v26.10)Mistral AIPreviewGPT-6 LunaOpenAImax
K2 Horizon 375B A23BInstitute of Foundation Models
MiniMax-M3MiniMax
Solar Pro 4Upstagelisted on AA as Solar Pro 45; asterisk on page
Hy3Tencent
Nemotron 3 Ultra 550B A55BNVIDIAReasoning
K-EXAONE 2.0LG AI Researchasterisk on page
Doubao Seed CodeByteDanceasterisk on page
Nova 2.0 OmniAmazonasterisk on pageNova 2.0 Pro PreviewAmazonmedium; asterisk on page
ERNIE 5.0 Thinking PreviewBaiduasterisk on pageMistral Medium 3.5 (v26.04)Mistral AI
Command A+ (command-a-plus-05-2026)Cohere
HyperCLOVA X SEED Think 32BNaverasterisk on page
Sarvam 105BSarvam AIhigh; asterisk on page
Jamba 1.7 LargeAI21 Labsasterisk on page
Claude Opus 5.5Anthropicmax; 57.62 per secondary orcarouter58 index
independent2Claude Sonnet 5.5Anthropicmax56 index
independent3Claude Fable 5.1Anthropicmax53 index
independent4Gemini 4 ArgonGoogle DeepMindhigh; 52.56 per secondary53 index
independent5GPT-6 AstraOpenAImax; 52.67 per secondary53 index
independent6GPT-6.1 SolOpenAImax52 index
independent7Muse Spark 1.3Metamax48 index
independent8Grok 4.7SpaceXAI (xAI)xhigh46 index
independent9MiMo-V2.6-ProXiaomi46 index
independent10Qwen3.8 MaxAlibaba (Qwen)090245 index
independent11GLM-5.3Z.ai (Zhipu)max45 index
independent12Kimi K3Moonshot AImax44 index
independent13Step 5 PreviewStepFun44 index
independent14Claude Haiku 5.5Anthropicmax43 index
independent15Gemini 3.8 FlashGoogle DeepMindhigh; 40.9341 index
independent16DeepSeek-V4.1-FlashDeepSeekmax39 index
independent17Mistral Large 4 (v26.10)Mistral AIPreview38 index
independent18GPT-6 LunaOpenAImax38 index
independent19K2 Horizon 375B A23BInstitute of Foundation Models31 index
independent20MiniMax-M3MiniMax29 index
independent21Solar Pro 4Upstagelisted on AA as Solar Pro 45; asterisk on page28 index
independent22Hy3Tencent25 index
independent23Nemotron 3 Ultra 550B A55BNVIDIAReasoning23 index
independent24K-EXAONE 2.0LG AI Researchasterisk on page20 index
independent25Doubao Seed CodeByteDanceasterisk on page17 index
independent26Nova 2.0 OmniAmazonasterisk on page14 index
independent27Nova 2.0 Pro PreviewAmazonmedium; asterisk on page14 index
independent28ERNIE 5.0 Thinking PreviewBaiduasterisk on page14 index
independent29Mistral Medium 3.5 (v26.04)Mistral AI14 index
independent30Command A+ (command-a-plus-05-2026)Cohere13 index
independent31HyperCLOVA X SEED Think 32BNaverasterisk on page11 index
independent32Sarvam 105BSarvam AIhigh; asterisk on page9 index
independent33Jamba 1.7 LargeAI21 Labsasterisk on page6 index
independent0204060
Score, AA model page, rank within model class
Higher is better for this unit.
1MiMo-V2.6-FlashXiaomiIntelligence Index v4.3.2MiMo-V2.5XiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on pageMiMo-V2-OmniXiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Omni-0327MiMo-V2-FlashXiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Flash (Feb 2026)
Olmo 3.1 32B InstructAllen AIIntelligence Index v4.3.2; marked estimated (asterisk) on page
MiMo-V2.6-FlashXiaomiIntelligence Index v4.3.238 index
independent2MiMo-V2.5XiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on page25 index
independent3MiMo-V2-OmniXiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Omni-032725 index
independent4MiMo-V2-FlashXiaomiIntelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Flash (Feb 2026)22 index
independent5Olmo 3.1 32B InstructAllen AIIntelligence Index v4.3.2; marked estimated (asterisk) on page6 index
independent0204060
Score, AA model page, 227 models
Higher is better for this unit.
1Claude Opus 5.5AnthropicmaxMuse Spark 1.3MetamaxGrok 4.7SpaceXAI (xAI)xhigh
Claude Opus 5.5Anthropicmax58 index
independent2Muse Spark 1.3Metamax48 index
independent3Grok 4.7SpaceXAI (xAI)xhigh46 index
independent0204060
Score, Artificial Analysis open-weights view
Higher is better for this unit.
1MiMo-V2.6-ProXiaomiGLM-5.3Z.ai (Zhipu)GLM-5.3 (Max)Kimi K3Moonshot AIKimi K3 (Max)
MiMo-V2.6-ProXiaomi46 index
independent2GLM-5.3Z.ai (Zhipu)GLM-5.3 (Max)45 index
independent3Kimi K3Moonshot AIKimi K3 (Max)44 index
independent0204060
Score, Artificial Analysis model page
Higher is better for this unit.
1Qwen3.8 MaxAlibaba (Qwen)Qwen3.8 Max (0902)Qwen3.8-2.4T-A95BAlibaba (Qwen)Qwen3.8 2.4T A95BMistral Large 4 (v26.10)Mistral AIMistral Large 4 Preview
Qwen3.8 MaxAlibaba (Qwen)Qwen3.8 Max (0902)45 index
independent2Qwen3.8-2.4T-A95BAlibaba (Qwen)Qwen3.8 2.4T A95B40 index
independent3Mistral Large 4 (v26.10)Mistral AIMistral Large 4 Preview38 index
independent0204060
Score, DataLearner mirror of AA Index v4.0
Higher is better for this unit.
1Gemini 3.1 Pro (preview)Google DeepMind, Googlev4.0, stale; superseded by AA v4.3.2GPT-5.4OpenAIGPT-5.4 (xhigh); v4.0, stale; superseded by AA v4.3.2
Gemini 3.1 Pro (preview)Google DeepMind, Googlev4.0, stale; superseded by AA v4.3.257 index
independent2GPT-5.4OpenAIGPT-5.4 (xhigh); v4.0, stale; superseded by AA v4.3.257 index
independent0204060
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 | 58 index | independent | max; 57.62 per secondary orcarouter | rank 1 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Claude Opus 5.5 | 58 index | independent | max | rank 1 of 227, AA model page, 227 models | 10 Oct 2026 | artificialanalysis.ai |
| Gemini 3.1 Pro (preview) | 57 index | independent | v4.0, stale; superseded by AA v4.3.2 | rank 1, DataLearner mirror of AA Index v4.0 | 26 Mar 2026 | datalearner.com |
| GPT-5.4 | 57 index | independent | GPT-5.4 (xhigh); v4.0, stale; superseded by AA v4.3.2 | rank 1, DataLearner mirror of AA Index v4.0 | 26 Mar 2026 | datalearner.com |
| Claude Sonnet 5.5 | 56 index | independent | max | rank 2 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Claude Fable 5.1 | 53 index | independent | max | rank 5 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Gemini 4 Argon | 53 index | independent | high; 52.56 per secondary | rank 8 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| GPT-6 Astra | 53 index | independent | max; 52.67 per secondary | rank 7 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| GPT-6.1 Sol | 52 index | independent | max | rank 11 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Muse Spark 1.3 | 48 index | independent | max | rank 19 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Muse Spark 1.3 | 48 index | independent | max | rank 23 of 227, AA model page, 227 models | 10 Oct 2026 | artificialanalysis.ai |
| Grok 4.7 | 46 index | independent | xhigh | rank 23 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Grok 4.7 | 46 index | independent | xhigh | rank 29 of 227, AA model page, 227 models | 10 Oct 2026 | artificialanalysis.ai |
| MiMo-V2.6-Pro | 46 index | independent | – | rank 25 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| MiMo-V2.6-Pro | 46 index | independent | – | rank 1, Artificial Analysis open-weights view | – | artificialanalysis.ai |
| Qwen3.8 Max | 45 index | independent | 0902 | rank 27 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Qwen3.8 Max | 45 index | independent | Qwen3.8 Max (0902) | Artificial Analysis model page | – | artificialanalysis.ai |
| GLM-5.3 | 45 index | independent | GLM-5.3 (Max) | rank 2, Artificial Analysis open-weights view | – | artificialanalysis.ai |
| GLM-5.3 | 45 index | independent | max | rank 29 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Kimi K3 | 44 index | independent | Kimi K3 (Max) | rank 3, Artificial Analysis open-weights view | – | artificialanalysis.ai |
| Kimi K3 | 44 index | independent | max | rank 31 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Step 5 Preview | 44 index | independent | – | rank 30 of 205, AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Claude Haiku 5.5 | 43 index | independent | max | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Gemini 3.8 Flash | 41 index | independent | high; 40.93 | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Qwen3.8-2.4T-A95B | 40 index | independent | – | AA model page | 10 Oct 2026 | artificialanalysis.ai |
| Qwen3.8-2.4T-A95B | 40 index | independent | Qwen3.8 2.4T A95B | Artificial Analysis model page | – | artificialanalysis.ai |
| DeepSeek-V4.1-Flash | 39 index | independent | max | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Mistral Large 4 (v26.10) | 38 index | independent | Mistral Large 4 Preview | Artificial Analysis model page | – | artificialanalysis.ai |
| Mistral Large 4 (v26.10) | 38 index | independent | Preview | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| GPT-6 Luna | 38 index | independent | max | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| MiMo-V2.6-Flash | 38 index | independent | Intelligence Index v4.3.2 | rank 8 of 117, AA model page, rank within model class | 11 Oct 2026 | artificialanalysis.ai |
| K2 Horizon 375B A23B | 31 index | independent | – | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| MiniMax-M3 | 29 index | independent | – | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Solar Pro 4 | 28 index | independent | listed on AA as Solar Pro 45; asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Hy3 | 25 index | independent | – | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| MiMo-V2.5 | 25 index | independent | Intelligence Index v4.3.2; marked estimated (asterisk) on page | rank 27 of 117, AA model page, rank within model class | 11 Oct 2026 | artificialanalysis.ai |
| MiMo-V2-Omni | 25 index | independent | Intelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Omni-0327 | rank 161 of 697, AA model page, rank within model class | 11 Oct 2026 | artificialanalysis.ai |
| Nemotron 3 Ultra 550B A55B | 23 index | independent | Nemotron 3 Ultra 550B A55B Reasoning | rank 36 of 117, Artificial Analysis open-weights model page | – | artificialanalysis.ai |
| Nemotron 3 Ultra 550B A55B | 23 index | independent | Reasoning | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Nemotron 3 Ultra 550B A55B | 23 index | independent | Reasoning | rank 36 of 117, AA open-weights view, 117 models | 10 Oct 2026 | artificialanalysis.ai |
| MiMo-V2-Flash | 22 index | independent | Intelligence Index v4.3.2; marked estimated (asterisk) on page; AA entry MiMo-V2-Flash (Feb 2026) | rank 40 of 117, AA model page, rank within model class | 11 Oct 2026 | artificialanalysis.ai |
| K-EXAONE 2.0 | 20 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Doubao Seed Code | 17 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Nova 2.0 Omni | 14 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Nova 2.0 Pro Preview | 14 index | independent | medium; asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| ERNIE 5.0 Thinking Preview | 14 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Mistral Medium 3.5 (v26.04) | 14 index | independent | – | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Command A+ (command-a-plus-05-2026) | 13 index | independent | – | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
 MiniCPM-V 4.6 1.3B | 13 index | independent | – | rank 1, Open models under 2B | 11 May 2026 | artificialanalysis.ai |
| HyperCLOVA X SEED Think 32B | 11 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Sarvam 105B | 9 index | independent | high; asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Jamba 1.7 Large | 6 index | independent | asterisk on page | AA leaderboard, 205 current models | 10 Oct 2026 | artificialanalysis.ai |
| Olmo 3.1 32B Instruct | 6 index | independent | Intelligence Index v4.3.2; marked estimated (asterisk) on page | rank 34 of 75, AA model page, rank within model class | 11 Oct 2026 | artificialanalysis.ai |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Artificial Analysis Intelligence Index", https://aurion.semai.id/benchmarks/aa-intelligence-index/, records of October 2026.