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 13 min ago · first seen 11 Sept 2026

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

Score history · gpt-5.6-luna 6 rows

Not enough history to chart — 6 observations, all dated 11 Sept 2026. Rows under different configurations count separately; the list below shows each one.

  • 21.55aa_slug=gpt-5-6-luna-low · version=4.3 · estimated=false · aa_variant_slug=gpt-5-6-luna-low11 Sept 2026
  • 37.5aa_slug=gpt-5-6-luna · version=4.3 · estimated=false11 Sept 2026
  • 25.48aa_slug=gpt-5-6-luna-medium · version=4.3 · estimated=false · aa_variant_slug=gpt-5-6-luna-medium11 Sept 2026
  • 34.77aa_slug=gpt-5-6-luna-xhigh · version=4.3 · estimated=false · aa_variant_slug=gpt-5-6-luna-xhigh11 Sept 2026
  • 32.41aa_slug=gpt-5-6-luna-high · version=4.3 · estimated=false · aa_variant_slug=gpt-5-6-luna-high11 Sept 2026
  • 16.76aa_slug=gpt-5-6-luna-non-reasoning · version=4.3 · estimated=true · reasoning=off11 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
301Llama 3.3 70BOpen weightsMeta AI · Llama 3.37.66Independentversion4.3Comparable0.00obs. 11 Sept 2026artificialanalysis.aiT2History
302qwen3-30b-a3b-instructOpen weightsAlibaba Group · Qwen3 · best of 2 rows7.63Independentreasoningonversion4.3Partially comparable-0.03obs. 11 Sept 2026artificialanalysis.aiT2History
303Sonar ProClosedPerplexity AI · Sonar7.61Independentversion4.3Comparable-0.05obs. 11 Sept 2026artificialanalysis.aiT2History
304devstral-smallOpen weightsMistral AI · Devstral7.60Independentversion4.3Comparable-0.06obs. 11 Sept 2026artificialanalysis.aiT2History
305QwQ-32B-PreviewOpen weightsAlibaba Group · Qwen7.59Independentversion4.3Comparable-0.07obs. 11 Sept 2026artificialanalysis.aiT2History
306GLM 4.5VOpen weightsZ.ai (Zhipu AI) · GLM4.5 · best of 2 rows7.56Independentversion4.3Comparable-0.10obs. 11 Sept 2026artificialanalysis.aiT2History
306mistral-large-2Open weightsMistral AI · Mistral7.56Independentversion4.3Comparable-0.10obs. 11 Sept 2026artificialanalysis.aiT2History
308llama-3-1-nemotron-ultra-253b-v1-reasoningOpen weightsNVIDIA · Llama 3.17.53Independentversion4.3Comparable-0.13obs. 11 Sept 2026artificialanalysis.aiT2History
309ernie-4-5-300b-a47bOpen weightsBaidu · ERNIE 4.57.50Independentversion4.3Comparable-0.16obs. 11 Sept 2026artificialanalysis.aiT2History
310Hermes 4 405BOpen weightsNous Research · Hermes 47.49Independentversion4.3Comparable-0.17obs. 11 Sept 2026artificialanalysis.aiT2History
310solar-pro-2ClosedUpstage · Solar · best of 2 rows7.49Independentreasoningonversion4.3Partially comparable-0.17obs. 11 Sept 2026artificialanalysis.aiT2History
312nvidia-nemotron-nano-12b-v2-vlOpen weightsNVIDIA · Nemotron · best of 2 rows7.48Independentreasoningonversion4.3Partially comparable-0.18obs. 11 Sept 2026artificialanalysis.aiT2History
313granite-4.1-30bOpen weightsIBM · Granite 4.17.44Independentversion4.3Comparable-0.22obs. 11 Sept 2026artificialanalysis.aiT2History
314NVIDIA-Nemotron-Nano-9B-v2Open weightsNVIDIA · Nemotron · best of 2 rows7.43Independentversion4.3Comparable-0.23obs. 11 Sept 2026artificialanalysis.aiT2History
315gemini-2-0-flash-lite-001ClosedGoogle · Gemini 2.07.41Independentversion4.3Comparable-0.25obs. 11 Sept 2026artificialanalysis.aiT2History
315hermes-4-llama-3-1-405bOpen weightsNous Research · Llama 3.17.41Independentversion4.3Comparable-0.25obs. 11 Sept 2026artificialanalysis.aiT2History
317nvidia-nemotron-3-nano-4bOpen weightsNVIDIA · Nemotron 37.38Independentversion4.3Comparable-0.28obs. 11 Sept 2026artificialanalysis.aiT2History
318Mistral Small 3.1Open weightsMistral AI · Mistral7.37Independentversion4.3Comparable-0.29obs. 11 Sept 2026artificialanalysis.aiT2History
319qwen3-32b-instructOpen weightsAlibaba Group · Qwen37.34Independentversion4.3Comparable-0.32obs. 11 Sept 2026artificialanalysis.aiT2History
320GPT-4o (2024-05-13)ClosedOpenAI · GPT 47.33Independentversion4.3Comparable-0.33obs. 11 Sept 2026artificialanalysis.aiT2History
320gemini-2-0-flash-lite-previewClosedGoogle · Gemini 2.07.33Independentversion4.3Comparable-0.33obs. 11 Sept 2026artificialanalysis.aiT2History
322llama-3-1-nemotron-nano-4b-reasoningOpen weightsNVIDIA · Llama 3.17.31Independentversion4.3Comparable-0.35obs. 11 Sept 2026artificialanalysis.aiT2History
323Kimi-Linear-48B-A3B-InstructOpen weightsMoonshot AI · Kimi7.30Independentversion4.3Comparable-0.36obs. 11 Sept 2026artificialanalysis.aiT2History
324Llama-3.1-405BRestricted weightsMeta AI · Llama 3.17.29Independentversion4.3Comparable-0.37obs. 11 Sept 2026artificialanalysis.aiT2History
325Qwen3-4BOpen weightsQwen · Qwen37.23Independentversion4.3Comparable-0.43obs. 11 Sept 2026artificialanalysis.aiT2History
326LFM2.5-8B-A1BOpen weightsLiquid AI · LFM2.57.22Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
326Qwen3 32BOpen weightsQwen · Qwen37.22Independentversion4.3Comparable-0.44obs. 11 Sept 2026artificialanalysis.aiT2History
328claude-35-sonnet-june-24ClosedAnthropic · Claude 357.21Independentversion4.3Comparable-0.45obs. 11 Sept 2026artificialanalysis.aiT2History
329tulu3-405bOpen weightsAllen Institute for AI7.20Independentversion4.3Comparable-0.46obs. 11 Sept 2026artificialanalysis.aiT2History
330gpt-4o-chatgptClosedOpenAI · GPT 47.19Independentversion4.3Comparable-0.47obs. 11 Sept 2026artificialanalysis.aiT2History
331Pixtral LargeOpen weightsMistral AI · Pixtral7.15Independentversion4.3Comparable-0.51obs. 11 Sept 2026artificialanalysis.aiT2History
331ring-flash-2-0Open weightsinclusionAI7.15Independentversion4.3Comparable-0.51obs. 11 Sept 2026artificialanalysis.aiT2History
333olmo-3-1-32b-instructOpen weightsAllen Institute for AI · OLMo 3.1 · best of 2 rows7.12Independentreasoningonversion4.3Partially comparable-0.54obs. 11 Sept 2026artificialanalysis.aiT2History
334grok-2Open weightsxAI · Grok 27.11Independentversion4.3Comparable-0.55obs. 11 Sept 2026artificialanalysis.aiT2History
335gemini-1-5-flashClosedGoogle · Gemini 1.57.07Independentversion4.3Comparable-0.59obs. 11 Sept 2026artificialanalysis.aiT2History
336Qwen3-VL-4B-InstructOpen weightsQwen · Qwen3 · best of 2 rows7.05Independentreasoningonversion4.3Partially comparable-0.61obs. 11 Sept 2026artificialanalysis.aiT2History
337gpt-4-turboClosedOpenAI · GPT 47.04Independentversion4.3Comparable-0.62obs. 11 Sept 2026artificialanalysis.aiT2History
338Command AOpen weightsCohere · Command6.96Independentversion4.3Comparable-0.70obs. 11 Sept 2026artificialanalysis.aiT2History
338Mistral Small 3.2Open weightsMistral AI · Mistral6.96Independentversion4.3Comparable-0.70obs. 11 Sept 2026artificialanalysis.aiT2History
338nova-proClosedAmazon Web Services · Nova6.96Independentversion4.3Comparable-0.70obs. 11 Sept 2026artificialanalysis.aiT2History
341Qwen3.5-2BOpen weightsQwen · Qwen3.5 · best of 2 rows6.94Independentversion4.3Comparable-0.72obs. 11 Sept 2026artificialanalysis.aiT2History
341llama-3-1-nemotron-instruct-70bOpen weightsNVIDIA · Llama 3.16.94Independentversion4.3Comparable-0.72obs. 11 Sept 2026artificialanalysis.aiT2History
343Llama 3.1 8BRestricted weightsMeta AI · Llama 3.16.93Independentversion4.3Comparable-0.73obs. 11 Sept 2026artificialanalysis.aiT2History
344grok-betaClosedSpaceXAI · Grok6.89Independentversion4.3Comparable-0.77obs. 11 Sept 2026artificialanalysis.aiT2History
345qwen2.5-32b-instructOpen weightsAlibaba Group · Qwen2.56.87Independentversion4.3Comparable-0.79obs. 11 Sept 2026artificialanalysis.aiT2History
346Mistral Large 2.0Open weightsMistral AI · Mistral6.80Independentversion4.3Comparable-0.86obs. 11 Sept 2026artificialanalysis.aiT2History
347Qwen2.5 Coder 32B InstructOpen weightsQwen · Qwen2.56.74Independentversion4.3Comparable-0.92obs. 11 Sept 2026artificialanalysis.aiT2History
348gpt-4ClosedOpenAI · GPT 46.70Independentversion4.3Comparable-0.96obs. 11 Sept 2026artificialanalysis.aiT2History
348qwen3-14b-instructOpen weightsAlibaba Group · Qwen36.70Independentversion4.3Comparable-0.96obs. 11 Sept 2026artificialanalysis.aiT2History
350Mistral Small 3Open weightsMistral AI · Mistral6.67Independentversion4.3Comparable-0.99obs. 11 Sept 2026artificialanalysis.aiT2History
350nova-liteClosedAmazon Web Services · Nova6.67Independentversion4.3Comparable-0.99obs. 11 Sept 2026artificialanalysis.aiT2History
352gpt-4o-miniClosedOpenAI · GPT 46.66Independentversion4.3Comparable-1.00obs. 11 Sept 2026artificialanalysis.aiT2History
352hermes-4-llama-3-1-70bOpen weightsNous Research · Llama 3.16.66Independentversion4.3Comparable-1.00obs. 11 Sept 2026artificialanalysis.aiT2History
354deepseek-v2-5Open weightsDeepSeek · DeepSeek6.62Independentversion4.3Comparable-1.04obs. 11 Sept 2026artificialanalysis.aiT2History
355qwen3-4b-instructOpen weightsAlibaba Group · Qwen36.61Independentversion4.3Comparable-1.05obs. 11 Sept 2026artificialanalysis.aiT2History
356Llama-3.1-70BOpen weightsMeta AI · Llama 3.16.60Independentversion4.3Comparable-1.06obs. 11 Sept 2026artificialanalysis.aiT2History
357granite-4.1-8bOpen weightsIBM · Granite 4.16.57Independentversion4.3Comparable-1.09obs. 11 Sept 2026artificialanalysis.aiT2History
358gemini-2-0-flash-thinking-exp-1219ClosedGoogle · Gemini 2.06.56Independentversion4.3Comparable-1.10obs. 11 Sept 2026artificialanalysis.aiT2History
358sarvam-30bOpen weightsSarvam6.56Independentversion4.3Comparable-1.10obs. 11 Sept 2026artificialanalysis.aiT2History
360deepseek-v2-5-sep-2024Open weightsDeepSeek · DeepSeek6.55Independentversion4.3Comparable-1.11obs. 11 Sept 2026artificialanalysis.aiT2History
361Mistral SabaClosedMistral AI · Mistral6.49Independentversion4.3Comparable-1.17obs. 11 Sept 2026artificialanalysis.aiT2History
362deepseek-r1-distill-llama-8bOpen weightsDeepSeek · Llama6.48Independentversion4.3Comparable-1.18obs. 11 Sept 2026artificialanalysis.aiT2History
363olmo-3-32b-thinkOpen weightsAllen Institute for AI · OLMo 36.47Independentversion4.3Comparable-1.19obs. 11 Sept 2026artificialanalysis.aiT2History
364Llama 4 ScoutOpen weightsMeta AI · Llama 46.45Independentversion4.3Comparable-1.21obs. 11 Sept 2026artificialanalysis.aiT2History
365gemini-1-5-pro-may-2024ClosedGoogle · Gemini 1.56.44Independentversion4.3Comparable-1.22obs. 11 Sept 2026artificialanalysis.aiT2History
365r1-1776Open weightsPerplexity AI6.44Independentversion4.3Comparable-1.22obs. 11 Sept 2026artificialanalysis.aiT2History
367qwen-turboClosedAlibaba Group · Qwen6.43Independentversion4.3Comparable-1.23obs. 11 Sept 2026artificialanalysis.aiT2History
367reka-flashClosedrekaai6.43Independentversion4.3Comparable-1.23obs. 11 Sept 2026artificialanalysis.aiT2History
369llama-3-2-instruct-90b-visionOpen weightsMeta AI · Llama 3.26.41Independentversion4.3Comparable-1.25obs. 11 Sept 2026artificialanalysis.aiT2History
369solar-miniOpen weightsUpstage · Solar6.41Independentversion4.3Comparable-1.25obs. 11 Sept 2026artificialanalysis.aiT2History
371Qwen3 14BOpen weightsQwen · Qwen36.39Independentversion4.3Comparable-1.27obs. 11 Sept 2026artificialanalysis.aiT2History
372celeris-1ClosedCeleris6.35Independentversion4.3Comparable-1.31obs. 11 Sept 2026artificialanalysis.aiT2History
373grok-1Open weightsxAI · Grok 16.34Independentversion4.3Comparable-1.32obs. 11 Sept 2026artificialanalysis.aiT2History
374phi-4-miniOpen weightsMicrosoft · Phi46.33Independentversion4.3Comparable-1.33obs. 11 Sept 2026artificialanalysis.aiT2History
374qwen2-72b-instructOpen weightsAlibaba Group · Qwen26.33Independentversion4.3Comparable-1.33obs. 11 Sept 2026artificialanalysis.aiT2History
376gemini-1-5-flash-8bClosedGoogle · Gemini 1.56.15Independentversion4.3Comparable-1.51obs. 11 Sept 2026artificialanalysis.aiT2History
377Qwen3.5-0.8BOpen weightsQwen · Qwen3.5 · best of 2 rows6.13Independentversion4.3Comparable-1.53obs. 11 Sept 2026artificialanalysis.aiT2History
378deephermes-3-mistral-24b-previewOpen weightsNous Research · Mistral6.07Independentversion4.3Comparable-1.59obs. 11 Sept 2026artificialanalysis.aiT2History
378jamba-1-7-largeOpen weightsAI21 Labs · Jamba 1.76.07Independentversion4.3Comparable-1.59obs. 11 Sept 2026artificialanalysis.aiT2History
380granite-4-0-h-smallOpen weightsIBM · Granite 4.06.05Independentversion4.3Comparable-1.61obs. 11 Sept 2026artificialanalysis.aiT2History
381Ministral 3 14BOpen weightsMistral AI · Ministral 36.04Independentversion4.3Comparable-1.62obs. 11 Sept 2026artificialanalysis.aiT2History
382jamba-1-5-largeOpen weightsAI21 Labs · Jamba 1.56.01Independentversion4.3Comparable-1.65obs. 11 Sept 2026artificialanalysis.aiT2History
383Hermes 3 70B InstructOpen weightsNous Research · Hermes 36Independentversion4.3Comparable-1.66obs. 11 Sept 2026artificialanalysis.aiT2History
384qwen3-8b-instructOpen weightsAlibaba Group · Qwen35.99Independentversion4.3Comparable-1.67obs. 11 Sept 2026artificialanalysis.aiT2History
385deepseek-coder-v2Open weightsDeepSeek · DeepSeek5.98Independentversion4.3Comparable-1.68obs. 11 Sept 2026artificialanalysis.aiT2History
386jamba-1-6-largeOpen weightsAI21 Labs · Jamba 1.65.97Independentversion4.3Comparable-1.69obs. 11 Sept 2026artificialanalysis.aiT2History
386olmo-2-32bOpen weightsAllen Institute for AI · OLMo 25.97Independentversion4.3Comparable-1.69obs. 11 Sept 2026artificialanalysis.aiT2History
388lfm2-24b-a2bOpen weightsLiquid AI · LFM25.95Independentversion4.3Comparable-1.71obs. 11 Sept 2026artificialanalysis.aiT2History
389gemini-1-5-flash-may-2024ClosedGoogle · Gemini 1.55.94Independentversion4.3Comparable-1.72obs. 11 Sept 2026artificialanalysis.aiT2History
390Phi 4Open weightsMicrosoft · Phi45.92Independentversion4.3Comparable-1.74obs. 11 Sept 2026artificialanalysis.aiT2History
391claude-3-sonnetClosedAnthropic · Claude 35.88Independentversion4.3Comparable-1.78obs. 11 Sept 2026artificialanalysis.aiT2History
391nova-microClosedAmazon Web Services · Nova5.88Independentversion4.3Comparable-1.78obs. 11 Sept 2026artificialanalysis.aiT2History
393granite-4.1-3bOpen weightsIBM · Granite 4.15.86Independentversion4.3Comparable-1.80obs. 11 Sept 2026artificialanalysis.aiT2History
394mistral-smallOpen weightsMistral AI · Mistral5.85Independentversion4.3Comparable-1.81obs. 11 Sept 2026artificialanalysis.aiT2History
395gemini-1-0-ultraClosedGoogle · Gemini 1.05.84Independentversion4.3Comparable-1.82obs. 11 Sept 2026artificialanalysis.aiT2History
396phi-3-miniOpen weightsMicrosoft · Phi35.82Independentversion4.3Comparable-1.84obs. 11 Sept 2026artificialanalysis.aiT2History
397gemma-3n-e4b-preview-0520Open weightsGoogle · Gemma 35.81Independentversion4.3Comparable-1.85obs. 11 Sept 2026artificialanalysis.aiT2History
397phi-4-multimodalOpen weightsMicrosoft · Phi45.81Independentversion4.3Comparable-1.85obs. 11 Sept 2026artificialanalysis.aiT2History
399Qwen2.5-Coder-7BOpen weightsQwen · Qwen2.55.79Independentversion4.3Comparable-1.87obs. 11 Sept 2026artificialanalysis.aiT2History
400Mistral LargeClosedMistral AI · Mistral5.76Independentversion4.3Comparable-1.90obs. 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 →