Skip to content
AI Atlas
BenchmarkActivecategory · compositefamily · artificial-analysis-intelligence-index · variant index

Artificial Analysis Intelligence Index

artificialanalysis.ai

composite of several evaluations run by Artificial Analysis

quality57

Updated 11 min ago · first seen 11 Sept 2026

Metric
index ↑
Current results
638
Models
477
Current leader
Claude Fable 5.1 53.4

Score history · Muse Spark 1.2 1 row

Not enough history to chart — a single observation (39.8 on 11 Sept 2026). Rows under different configurations count separately; the list below shows each one.

  • 39.8aa_slug=muse-spark-1-2 · version=4.3 · estimated=false11 Sept 2026

Back to the leaderboard

Frontier over time · index

8 leader changes recorded, all dated 11 Sept 2026 — the frontier line needs at least two distinct dates. The corpus is young: every result was first observed on the same day, so leader changes will separate in time as sources are re-crawled.

  1. 53.4Claude Fable 5.1 Anthropic Independent11 Sept 2026
  2. 53.2Claude Fable 5.1 Anthropic Independent11 Sept 2026
  3. 51.2Claude Fable 5.1 Anthropic Independent11 Sept 2026
  4. 51.0gpt-6-astra OpenAI Independent11 Sept 2026
  5. 49.7gpt-6-astra OpenAI Independent11 Sept 2026
  6. 45.1Claude Opus 5 Anthropic Independent11 Sept 2026
  7. 14.6grok-3-mini-reasoning SpaceXAI Independent11 Sept 2026
  8. 5.71llama-2-chat-7b Meta AI Independent11 Sept 2026

Includes closed rows (history). A point is emitted whenever a result beats every earlier result of the same group, ordered by evaluated_at when the source publishes it, else observed_at.

Leaderboard 477 models

Select models with +, then Compare.

Leaderboard
#ModelScoreTrustConfigurationvs leaderEvaluatedSourceActions
401Mixtral 8x22BOpen weightsMistral AI · Mixtral 85.74Independentversion4.3Comparable0.00obs. 11 Sept 2026artificialanalysis.aiT2History
402llama-2-chat-7bOpen weightsMeta AI · Llama 25.71Independentversion4.3Comparable-0.03obs. 11 Sept 2026artificialanalysis.aiT2History
403Llama-3.2-3BRestricted weightsMeta AI · Llama 3.25.70Independentversion4.3Comparable-0.04obs. 11 Sept 2026artificialanalysis.aiT2History
404minicpm-v4-6-1-3bOpen weightsOpenBMB5.69Independentversion4.3Comparable-0.05obs. 11 Sept 2026artificialanalysis.aiT2History
405jamba-reasoning-3bOpen weightsAI21 Labs · Jamba5.67Independentversion4.3Comparable-0.07obs. 11 Sept 2026artificialanalysis.aiT2History
406Reka Flash 3Open weightsrekaai5.65Independentversion4.3Comparable-0.09obs. 11 Sept 2026artificialanalysis.aiT2History
406qwen1.5-110b-chatOpen weightsAlibaba Group · Qwen1.55.65Independentversion4.3Comparable-0.09obs. 11 Sept 2026artificialanalysis.aiT2History
408Olmo-3-7B-ThinkOpen weightsAllen Institute for AI · OLMo 35.62Independentversion4.3Comparable-0.12obs. 11 Sept 2026artificialanalysis.aiT2History
409claude-21ClosedAnthropic · Claude 215.59Independentversion4.3Comparable-0.15obs. 11 Sept 2026artificialanalysis.aiT2History
410Claude 3 HaikuClosedAnthropic · Claude5.58Independentversion4.3Comparable-0.16obs. 11 Sept 2026artificialanalysis.aiT2History
410olmo-2-7bOpen weightsAllen Institute for AI · OLMo 25.58Independentversion4.3Comparable-0.16obs. 11 Sept 2026artificialanalysis.aiT2History
412molmo-7b-dOpen weightsAllen Institute for AI · Molmo5.56Independentversion4.3Comparable-0.18obs. 11 Sept 2026artificialanalysis.aiT2History
413ling-mini-2-0Open weightsinclusionAI5.54Independentversion4.3Comparable-0.20obs. 11 Sept 2026artificialanalysis.aiT2History
414DeepSeek-R1-Distill-Qwen-1.5BOpen weightsDeepSeek · Qwen5.51Independentversion4.3Comparable-0.23obs. 11 Sept 2026artificialanalysis.aiT2History
414claude-2ClosedAnthropic · Claude 25.51Independentversion4.3Comparable-0.23obs. 11 Sept 2026artificialanalysis.aiT2History
414deepseek-v2Open weightsDeepSeek · DeepSeek5.51Independentversion4.3Comparable-0.23obs. 11 Sept 2026artificialanalysis.aiT2History
417Mistral Small 1.0ClosedMistral AI · Mistral5.50Independentversion4.3Comparable-0.24obs. 11 Sept 2026artificialanalysis.aiT2History
418gpt-3.5-turboClosedOpenAI · GPT 3.55.49Independentversion4.3Comparable-0.25obs. 11 Sept 2026artificialanalysis.aiT2History
418mistral-mediumClosedMistral AI · Mistral5.49Independentversion4.3Comparable-0.25obs. 11 Sept 2026artificialanalysis.aiT2History
420Ministral 3 8BOpen weightsMistral AI · Ministral 35.48Independentversion4.3Comparable-0.26obs. 11 Sept 2026artificialanalysis.aiT2History
421llama-3-instruct-70bOpen weightsMeta AI · Llama 35.45Independentversion4.3Comparable-0.29obs. 11 Sept 2026artificialanalysis.aiT2History
422arctic-instructOpen weightsSnowflake5.44Independentversion4.3Comparable-0.30obs. 11 Sept 2026artificialanalysis.aiT2History
422qwen-chat-72bOpen weightsAlibaba Group · Qwen5.44Independentversion4.3Comparable-0.30obs. 11 Sept 2026artificialanalysis.aiT2History
424lfm-40bClosedLiquid AI · LFM5.42Independentversion4.3Comparable-0.32obs. 11 Sept 2026artificialanalysis.aiT2History
425llama-3-2-instruct-11b-visionOpen weightsMeta AI · Llama 3.25.41Independentversion4.3Comparable-0.33obs. 11 Sept 2026artificialanalysis.aiT2History
426palm-2ClosedGoogle · PaLM 25.37Independentversion4.3Comparable-0.37obs. 11 Sept 2026artificialanalysis.aiT2History
427deepseek-coder-v2-liteOpen weightsDeepSeek · DeepSeek5.34Independentversion4.3Comparable-0.40obs. 11 Sept 2026artificialanalysis.aiT2History
427gemini-1-0-proClosedGoogle · Gemini 1.05.34Independentversion4.3Comparable-0.40obs. 11 Sept 2026artificialanalysis.aiT2History
429sarvam-m-reasoningOpen weightsSarvam5.31Independentversion4.3Comparable-0.43obs. 11 Sept 2026artificialanalysis.aiT2History
430command-r-plus-04-2024Open weightsCohere · Command5.30Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
430deepseek-llm-67b-chatOpen weightsDeepSeek · DeepSeek5.30Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
430llama-2-chat-13bOpen weightsMeta AI · Llama 25.30Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
430llama-2-chat-70bOpen weightsMeta AI · Llama 25.30Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
434dbrxOpen weightsDatabricks5.29Independentversion4.3Comparable-0.45obs. 11 Sept 2026artificialanalysis.aiT2History
434openchat-35Open weightsOpenChat5.29Independentversion4.3Comparable-0.45obs. 11 Sept 2026artificialanalysis.aiT2History
436exaone-4-0-1-2bOpen weightsLG AI Research · EXAONE 4.0 · best of 2 rows5.27Independentreasoningonversion4.3Partially comparable-0.47obs. 11 Sept 2026artificialanalysis.aiT2History
437Olmo-3-7B-InstructOpen weightsAllen Institute for AI · OLMo 35.24Independentversion4.3Comparable-0.50obs. 11 Sept 2026artificialanalysis.aiT2History
438jamba-1-7-miniOpen weightsAI21 Labs · Jamba 1.75.22Independentversion4.3Comparable-0.52obs. 11 Sept 2026artificialanalysis.aiT2History
438lfm2-5-1-2b-thinkingOpen weightsLiquid AI · LFM2.55.22Independentversion4.3Comparable-0.52obs. 11 Sept 2026artificialanalysis.aiT2History
440LFM2.5-1.2B-InstructOpen weightsLiquid AI · LFM2.55.21Independentversion4.3Comparable-0.53obs. 11 Sept 2026artificialanalysis.aiT2History
440jamba-1-5-miniOpen weightsAI21 Labs · Jamba 1.55.21Independentversion4.3Comparable-0.53obs. 11 Sept 2026artificialanalysis.aiT2History
440lfm2-2-6bOpen weightsLiquid AI · LFM2.25.21Independentversion4.3Comparable-0.53obs. 11 Sept 2026artificialanalysis.aiT2History
443granite-4-0-h-nano-1bOpen weightsIBM · Granite 4.05.19Independentversion4.3Comparable-0.55obs. 11 Sept 2026artificialanalysis.aiT2History
443qwen3-1.7b-instructOpen weightsAlibaba Group · Qwen3.1 · best of 2 rows5.19Independentreasoningonversion4.3Partially comparable-0.55obs. 11 Sept 2026artificialanalysis.aiT2History
445Qwen3 8BOpen weightsQwen · Qwen35.16Independentversion4.3Comparable-0.58obs. 11 Sept 2026artificialanalysis.aiT2History
445jamba-1-6-miniOpen weightsAI21 Labs · Jamba 1.65.16Independentversion4.3Comparable-0.58obs. 11 Sept 2026artificialanalysis.aiT2History
447Mixtral 8x7BOpen weightsMistral AI · Mixtral 85.12Independentversion4.3Comparable-0.62obs. 11 Sept 2026artificialanalysis.aiT2History
447gemma-3-270mOpen weightsGoogle · Gemma 35.12Independentversion4.3Comparable-0.62obs. 11 Sept 2026artificialanalysis.aiT2History
449Granite 4.0 MicroOpen weightsIBM · Granite 4.05.11Independentversion4.3Comparable-0.63obs. 11 Sept 2026artificialanalysis.aiT2History
449apertus-70b-instructOpen weightsSwiss AI Initiative5.11Independentversion4.3Comparable-0.63obs. 11 Sept 2026artificialanalysis.aiT2History
451deephermes-3-llama-3-1-8b-previewOpen weightsNous Research · Llama 3.15.08Independentversion4.3Comparable-0.66obs. 11 Sept 2026artificialanalysis.aiT2History
452Mistral 7BOpen weightsMistral AI · Mistral5.03Independentversion4.3Comparable-0.71obs. 11 Sept 2026artificialanalysis.aiT2History
452claude-instantClosedAnthropic · Claude5.03Independentversion4.3Comparable-0.71obs. 11 Sept 2026artificialanalysis.aiT2History
452command-r-03-2024Open weightsCohere · Command5.03Independentversion4.3Comparable-0.71obs. 11 Sept 2026artificialanalysis.aiT2History
452llama-65bOpen weightsMeta AI · Llama5.03Independentversion4.3Comparable-0.71obs. 11 Sept 2026artificialanalysis.aiT2History
452qwen-chat-14bOpen weightsAlibaba Group · Qwen5.03Independentversion4.3Comparable-0.71obs. 11 Sept 2026artificialanalysis.aiT2History
457granite-4-0-nano-1bOpen weightsIBM · Granite 4.05.01Independentversion4.3Comparable-0.73obs. 11 Sept 2026artificialanalysis.aiT2History
458Molmo2-8BOpen weightsAllen Institute for AI5Independentversion4.3Comparable-0.74obs. 11 Sept 2026artificialanalysis.aiT2History
459lfm2-8b-a1bOpen weightsLiquid AI · LFM24.93Independentversion4.3Comparable-0.81obs. 11 Sept 2026artificialanalysis.aiT2History
460granite-3-3-8b-instructOpen weightsIBM · Granite 3.34.92Independentversion4.3Comparable-0.82obs. 11 Sept 2026artificialanalysis.aiT2History
461Gemma 3 27BOpen weightsGoogle · Gemma 34.85Independentversion4.3Comparable-0.89obs. 11 Sept 2026artificialanalysis.aiT2History
462Ministral 3 3BOpen weightsMistral AI · Ministral 34.84Independentversion4.3Comparable-0.90obs. 11 Sept 2026artificialanalysis.aiT2History
463Gemma 3 4BRestricted weightsGoogle · Gemma 34.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463LFM2-1.2BOpen weightsLiquid AI · LFM2.14.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463LFM2.5-VL-1.6BOpen weightsLiquid AI · LFM2.54.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463Llama-3.2-1BRestricted weightsMeta AI · Llama 3.24.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463Qwen3-0.6BOpen weightsQwen · Qwen3.04.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463apertus-8b-instructOpen weightsSwiss AI Initiative4.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463gemma-3-1bOpen weightsGoogle · Gemma 34.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463gemma-3n-e2bOpen weightsGoogle · Gemma 34.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463gemma-3n-e4bOpen weightsGoogle · Gemma 34.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463granite-4-0-350mOpen weightsIBM · Granite 4.04.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463granite-4-0-h-350mOpen weightsIBM · Granite 4.04.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463llama-3-instruct-8bOpen weightsMeta AI · Llama 34.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463qwen3-0.6b-instructOpen weightsAlibaba Group · Qwen3.04.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
463tiny-aya-globalRestricted weightsCohere · Aya4.83Independentversion4.3Comparable-0.91obs. 11 Sept 2026artificialanalysis.aiT2History
477Gemma 3 12BOpen weightsGoogle · Gemma 33.75Independentversion4.3Comparable-1.99obs. 11 Sept 2026artificialanalysis.aiT2History

One row per canonical model — its best current row inside this comparability group (effort variants are folded into the model). Bars are relative to the page's best score. “vs leader” reads comparability: partially comparable = same task, conditions differ (reasoning effort, temperature, judge). Rules →