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 · qwen3-4b-2507-instruct 1 row

Not enough history to chart — a single observation (6.74 on 11 Sept 2026). Rows under different configurations count separately; the list below shows each one.

  • 6.74aa_slug=qwen3-4b-2507-instruct · version=4.3 · estimated=true11 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
201Qwen3 VL 32B InstructOpen weightsQwen · Qwen3 · best of 2 rows11.9Independentreasoningonversion4.3Partially comparable0.00obs. 11 Sept 2026artificialanalysis.aiT2History
201ling-3-0-tinyOpen weightsinclusionAI11.9Independentversion4.3Comparable0.00obs. 11 Sept 2026artificialanalysis.aiT2History
203Magistral Medium 1.2ClosedMistral AI · Magistral11.8Independentversion4.3Comparable-0.04obs. 11 Sept 2026artificialanalysis.aiT2History
204Sonar Reasoning ProClosedPerplexity AI · Sonar11.8Independentversion4.3Comparable-0.05obs. 11 Sept 2026artificialanalysis.aiT2History
205Granite 4.2 8BOpen weightsIBM · Granite 4.211.8Independentversion4.3Comparable-0.06obs. 11 Sept 2026artificialanalysis.aiT2History
206hypernova-60bOpen weightsMultiverse Computing11.7Independentversion4.3Comparable-0.13obs. 11 Sept 2026artificialanalysis.aiT2History
207MiniMax-M1-80kOpen weightsMiniMax · MiniMax11.7Independentversion4.3Comparable-0.14obs. 11 Sept 2026artificialanalysis.aiT2History
208nemotron-cascade-2-30b-a3bOpen weightsNVIDIA · Nemotron11.7Independentversion4.3Comparable-0.20obs. 11 Sept 2026artificialanalysis.aiT2History
209gemini-2-5-flash-reasoning-04-2025ClosedGoogle · Gemini 2.511.7Independentversion4.3Comparable-0.22obs. 11 Sept 2026artificialanalysis.aiT2History
210Mercury 2ClosedInception11.5Independentversion4.3Comparable-0.36obs. 11 Sept 2026artificialanalysis.aiT2History
211k2-think-v2Open weightsMBZUAI Institute of Foundation Models11.5Independentversion4.3Comparable-0.37obs. 11 Sept 2026artificialanalysis.aiT2History
212longcat-flash-liteOpen weightsLongCat11.5Independentversion4.3Comparable-0.40obs. 11 Sept 2026artificialanalysis.aiT2History
213Mistral Small 4Open weightsMistral AI · Mistral · best of 2 rows11.4Independentversion4.3Comparable-0.42obs. 11 Sept 2026artificialanalysis.aiT2History
214deepseek-r1-0120Open weightsDeepSeek · DeepSeek11.4Independentversion4.3Comparable-0.46obs. 11 Sept 2026artificialanalysis.aiT2History
215o1 PreviewClosedOpenAI · OpenAI o-series11.4Independentversion4.3Comparable-0.49obs. 11 Sept 2026artificialanalysis.aiT2History
216hyperclova-x-seed-think-32bOpen weightsNaver · Seed11.4Independentversion4.3Comparable-0.51obs. 11 Sept 2026artificialanalysis.aiT2History
217GLM 4.6VOpen weightsZ.ai (Zhipu AI) · GLM4.6 · best of 2 rows11.2Independentversion4.3Comparable-0.65obs. 11 Sept 2026artificialanalysis.aiT2History
218Qwen3 Next 80B A3B InstructOpen weightsQwen · Qwen3 · best of 2 rows11.2Independentreasoningonversion4.3Partially comparable-0.67obs. 11 Sept 2026artificialanalysis.aiT2History
219GLM 4.5 AirOpen weightsZ.ai (Zhipu AI) · GLM4.511.1Independentversion4.3Comparable-0.78obs. 11 Sept 2026artificialanalysis.aiT2History
220mi-dm-k-2-5-pro-dec28ClosedKorea Telecom11.0Independentversion4.3Comparable-0.83obs. 11 Sept 2026artificialanalysis.aiT2History
221ring-1tOpen weightsinclusionAI10.9Independentversion4.3Comparable-0.97obs. 11 Sept 2026artificialanalysis.aiT2History
222Trinity Large ThinkingOpen weightsArcee AI10.9Independentversion4.3Comparable-0.99obs. 11 Sept 2026artificialanalysis.aiT2History
223g9v3-3bOpen weightsAI9Stars10.8Independentversion4.3Comparable-1.03obs. 11 Sept 2026artificialanalysis.aiT2History
224intellect-3Open weightsPrime Intellect10.6Independentversion4.3Comparable-1.26obs. 11 Sept 2026artificialanalysis.aiT2History
225gpt-5-chatgptClosedOpenAI · GPT 510.4Independentversion4.3Comparable-1.43obs. 11 Sept 2026artificialanalysis.aiT2History
226solar-open-100b-reasoningOpen weightsUpstage · Solar10.4Independentversion4.3Comparable-1.51obs. 11 Sept 2026artificialanalysis.aiT2History
227gemini-2-5-flash-lite-preview-09-2025ClosedGoogle · Gemini 2.5 · best of 2 rows10.3Independentreasoningonversion4.3Partially comparable-1.52obs. 11 Sept 2026artificialanalysis.aiT2History
228nemotron-3-nano-omni-30b-a3bOpen weightsNVIDIA · Nemotron 310.3Independentversion4.3Comparable-1.62obs. 11 Sept 2026artificialanalysis.aiT2History
229gpt-4.1-miniClosedOpenAI · GPT 4.110.2Independentversion4.3Comparable-1.71obs. 11 Sept 2026artificialanalysis.aiT2History
230Qwen3 Coder NextOpen weightsQwen · Qwen310.1Independentversion4.3Comparable-1.82obs. 11 Sept 2026artificialanalysis.aiT2History
231MiniMax-M1-40kOpen weightsMiniMax · MiniMax9.99Independentversion4.3Comparable-1.88obs. 11 Sept 2026artificialanalysis.aiT2History
232gpt-oss-20bOpen weightsOpenAI · gpt-oss · best of 2 rows9.95Independentreasoning_effortlowversion4.3Partially comparable-1.92obs. 11 Sept 2026artificialanalysis.aiT2History
233Mistral Medium 3.1ClosedMistral AI · Mistral9.89Independentversion4.3Comparable-1.98obs. 11 Sept 2026artificialanalysis.aiT2History
234k2-v2Open weightsMBZUAI Institute of Foundation Models · best of 3 rows9.87Independentversion4.3Comparable-2.00obs. 11 Sept 2026artificialanalysis.aiT2History
235qwen3-30b-a3b-2507Open weightsAlibaba Group · Qwen3 · best of 2 rows9.81Independentreasoningonversion4.3Partially comparable-2.06obs. 11 Sept 2026artificialanalysis.aiT2History
236o1-miniClosedOpenAI · OpenAI o-series9.77Independentversion4.3Comparable-2.10obs. 11 Sept 2026artificialanalysis.aiT2History
237DeepSeek V3 0324Open weightsDeepSeek · DeepSeek-V39.72Independentversion4.3Comparable-2.15obs. 11 Sept 2026artificialanalysis.aiT2History
238Mistral Large 3Open weightsMistral AI · Mistral9.71Independentversion4.3Comparable-2.16obs. 11 Sept 2026artificialanalysis.aiT2History
239ling-2-6-flashOpen weightsinclusionAI9.68Independentversion4.3Comparable-2.19obs. 11 Sept 2026artificialanalysis.aiT2History
240Qwen3 Coder 30B A3B InstructOpen weightsQwen · Qwen39.59Independentversion4.3Comparable-2.28obs. 11 Sept 2026artificialanalysis.aiT2History
240tri-21b-think-previewOpen weightsTrillion Labs9.59Independentversion4.3Comparable-2.28obs. 11 Sept 2026artificialanalysis.aiT2History
242GPT-4.5 PreviewClosedOpenAI · GPT 4.59.58Independentversion4.3Comparable-2.29obs. 11 Sept 2026artificialanalysis.aiT2History
243diffusiongemma-26b-a4bOpen weightsGoogle9.51Independentversion4.3Comparable-2.36obs. 11 Sept 2026artificialanalysis.aiT2History
244qwen3-235b-a22b-instructOpen weightsAlibaba Group · Qwen3 · best of 2 rows9.50Independentreasoningonversion4.3Partially comparable-2.37obs. 11 Sept 2026artificialanalysis.aiT2History
245QwQ-32BOpen weightsAlibaba Group · Qwen9.47Independentversion4.3Comparable-2.40obs. 11 Sept 2026artificialanalysis.aiT2History
246Qwen3 VL 30B A3B InstructOpen weightsQwen · Qwen3 · best of 2 rows9.45Independentreasoningonversion4.3Partially comparable-2.42obs. 11 Sept 2026artificialanalysis.aiT2History
247gemini-2.0-flash-thinking-exp-01-21ClosedGoogle · Gemini 2.09.42Independentversion4.3Comparable-2.45obs. 11 Sept 2026artificialanalysis.aiT2History
248Devstral 2Open weightsMistral AI · Devstral 29.41Independentversion4.3Comparable-2.46obs. 11 Sept 2026artificialanalysis.aiT2History
249Llama 4 MaverickOpen weightsMeta AI · Llama 49.30Independentversion4.3Comparable-2.57obs. 11 Sept 2026artificialanalysis.aiT2History
250motif-2-12-7bClosedMotif Technologies9.19Independentversion4.3Comparable-2.68obs. 11 Sept 2026artificialanalysis.aiT2History
251ling-1tOpen weightsinclusionAI9.17Independentversion4.3Comparable-2.70obs. 11 Sept 2026artificialanalysis.aiT2History
252nova-premierClosedAmazon Web Services · Nova9.16Independentversion4.3Comparable-2.71obs. 11 Sept 2026artificialanalysis.aiT2History
253solar-pro-2-previewClosedUpstage · Solar · best of 2 rows9.07Independentreasoningonversion4.3Partially comparable-2.80obs. 11 Sept 2026artificialanalysis.aiT2History
254granite-4.2-3bOpen weightsIBM · Granite 4.29.06Independentversion4.3Comparable-2.81obs. 11 Sept 2026artificialanalysis.aiT2History
255Mistral Medium 3ClosedMistral AI · Mistral9.05Independentversion4.3Comparable-2.82obs. 11 Sept 2026artificialanalysis.aiT2History
255magistral-mediumClosedMistral AI · Magistral9.05Independentversion4.3Comparable-2.82obs. 11 Sept 2026artificialanalysis.aiT2History
257devstral-mediumClosedMistral AI · Devstral9.01Independentversion4.3Comparable-2.86obs. 11 Sept 2026artificialanalysis.aiT2History
257llama-nemotron-super-49b-v1-5Open weightsNVIDIA · Llama · best of 2 rows9.01Independentreasoningonversion4.3Partially comparable-2.86obs. 11 Sept 2026artificialanalysis.aiT2History
259tri-21b-think-v0-5Open weightsTrillion Labs8.99Independentversion4.3Comparable-2.88obs. 11 Sept 2026artificialanalysis.aiT2History
260gpt-4o-chatgpt-03-25ClosedOpenAI · GPT 48.96Independentversion4.3Comparable-2.91obs. 11 Sept 2026artificialanalysis.aiT2History
261Claude Haiku 3.5ClosedAnthropic · Claude8.94Independentversion4.3Comparable-2.93obs. 11 Sept 2026artificialanalysis.aiT2History
261Gemini 2.0 FlashClosedGoogle · Gemini8.94Independentversion4.3Comparable-2.93obs. 11 Sept 2026artificialanalysis.aiT2History
263llama-3-3-nemotron-super-49bOpen weightsNVIDIA · Llama 3.3 · best of 2 rows8.93Independentreasoningonversion4.3Partially comparable-2.94obs. 11 Sept 2026artificialanalysis.aiT2History
264gemma-4-E4BOpen weightsGoogle · Gemma 4 · best of 2 rows8.91Independentversion4.3Comparable-2.96obs. 11 Sept 2026artificialanalysis.aiT2History
265Nemotron 3 Nano 30B A3BOpen weightsNVIDIA · Nemotron 3 · best of 2 rows8.90Independentreasoningonversion4.3Partially comparable-2.97obs. 11 Sept 2026artificialanalysis.aiT2History
266minicpm5-1bOpen weightsOpenBMB · best of 2 rows8.80Independentversion4.3Comparable-3.07obs. 11 Sept 2026artificialanalysis.aiT2History
266qwen3-4b-2507-instructOpen weightsAlibaba Group · Qwen3 · best of 2 rows8.80Independentreasoningonversion4.3Partially comparable-3.07obs. 11 Sept 2026artificialanalysis.aiT2History
268sarvam-105bOpen weightsSarvam8.79Independentversion4.3Comparable-3.08obs. 11 Sept 2026artificialanalysis.aiT2History
269gemini-2-0-pro-experimental-02-05ClosedGoogle · Gemini 2.08.75Independentversion4.3Comparable-3.12obs. 11 Sept 2026artificialanalysis.aiT2History
270Devstral Small 1.0Open weightsMistral AI · Devstral8.74Independentversion4.3Comparable-3.13obs. 11 Sept 2026artificialanalysis.aiT2History
271claude-3-opusClosedAnthropic · Claude 38.72Independentversion4.3Comparable-3.15obs. 11 Sept 2026artificialanalysis.aiT2History
272SonarClosedPerplexity AI · Sonar · best of 2 rows8.67Independentreasoningonversion4.3Partially comparable-3.20obs. 11 Sept 2026artificialanalysis.aiT2History
273gemini-2-5-flash-04-2025ClosedGoogle · Gemini 2.58.66Independentversion4.3Comparable-3.21obs. 11 Sept 2026artificialanalysis.aiT2History
274Magistral Small 1.2Open weightsMistral AI · Magistral8.59Independentversion4.3Comparable-3.28obs. 11 Sept 2026artificialanalysis.aiT2History
275Gemini 2.5 Flash-LiteClosedGoogle · Gemini · best of 2 rows8.55Independentversion4.3Comparable-3.32obs. 11 Sept 2026artificialanalysis.aiT2History
276gpt-4oClosedOpenAI · GPT 48.44Independentversion4.3Comparable-3.43obs. 11 Sept 2026artificialanalysis.aiT2History
277nanbeige4-1-3bOpen weightsNanbeige8.40Independentversion4.3Comparable-3.47obs. 11 Sept 2026artificialanalysis.aiT2History
278LFM2.5-2.6B (free)Open weightsLiquid AI · LFM2.58.39Independentversion4.3Comparable-3.48obs. 11 Sept 2026artificialanalysis.aiT2History
279DeepSeek-R1-Distill-Qwen-32BOpen weightsDeepSeek · Qwen8.38Independentversion4.3Comparable-3.49obs. 11 Sept 2026artificialanalysis.aiT2History
280Devstral Small 2Open weightsMistral AI · Devstral8.36Independentversion4.3Comparable-3.51obs. 11 Sept 2026artificialanalysis.aiT2History
281gemini-2.0-flash-expClosedGoogle · Gemini 2.08.22Independentversion4.3Comparable-3.65obs. 11 Sept 2026artificialanalysis.aiT2History
281magistral-smallOpen weightsMistral AI · Magistral8.22Independentversion4.3Comparable-3.65obs. 11 Sept 2026artificialanalysis.aiT2History
283exaone-4-0-32bOpen weightsLG AI Research · EXAONE 4.0 · best of 2 rows8.18Independentreasoningonversion4.3Partially comparable-3.69obs. 11 Sept 2026artificialanalysis.aiT2History
284Qwen3 VL 8B InstructOpen weightsQwen · Qwen3 · best of 2 rows8.17Independentreasoningonversion4.3Partially comparable-3.70obs. 11 Sept 2026artificialanalysis.aiT2History
285DeepSeek-R1-0528-Qwen3-8BOpen weightsDeepSeek · Qwen38.08Independentversion4.3Comparable-3.79obs. 11 Sept 2026artificialanalysis.aiT2History
286qwen-2-5-maxClosedAlibaba Group · Qwen8.02Independentversion4.3Comparable-3.85obs. 11 Sept 2026artificialanalysis.aiT2History
287Hermes-4-70BRestricted weightsNous Research · Hermes 47.91Independentversion4.3Comparable-3.96obs. 11 Sept 2026artificialanalysis.aiT2History
287gemini-1-5-proClosedGoogle · Gemini 1.57.91Independentversion4.3Comparable-3.96obs. 11 Sept 2026artificialanalysis.aiT2History
289R1 Distill Llama 70BOpen weightsDeepSeek · Llama7.89Independentversion4.3Comparable-3.98obs. 11 Sept 2026artificialanalysis.aiT2History
290claude-35-sonnetClosedAnthropic · Claude 357.88Independentversion4.3Comparable-3.99obs. 11 Sept 2026artificialanalysis.aiT2History
291deepseek-r1-distill-qwen-14bOpen weightsDeepSeek · Qwen7.85Independentversion4.3Comparable-4.02obs. 11 Sept 2026artificialanalysis.aiT2History
292falcon-h1r-7bOpen weightsTII UAE · Falcon7.83Independentversion4.3Comparable-4.04obs. 11 Sept 2026artificialanalysis.aiT2History
293Solar Pro 3ClosedUpstage · Solar7.82Independentversion4.3Comparable-4.05obs. 11 Sept 2026artificialanalysis.aiT2History
293gpt-4.1-nanoClosedOpenAI · GPT 4.17.82Independentversion4.3Comparable-4.05obs. 11 Sept 2026artificialanalysis.aiT2History
295ling-flash-2-0Open weightsinclusionAI7.81Independentversion4.3Comparable-4.06obs. 11 Sept 2026artificialanalysis.aiT2History
296gemma-4-E2BOpen weightsGoogle · Gemma 4 · best of 2 rows7.77Independentversion4.3Comparable-4.10obs. 11 Sept 2026artificialanalysis.aiT2History
297qwen3-omni-30b-a3b-instructOpen weightsAlibaba Group · Qwen3 · best of 2 rows7.76Independentreasoningonversion4.3Partially comparable-4.11obs. 11 Sept 2026artificialanalysis.aiT2History
298GPT-4o (2024-08-06)ClosedOpenAI · GPT 47.74Independentversion4.3Comparable-4.13obs. 11 Sept 2026artificialanalysis.aiT2History
299Qwen2.5 72B InstructOpen weightsQwen · Qwen2.57.73Independentversion4.3Comparable-4.14obs. 11 Sept 2026artificialanalysis.aiT2History
300step-3-vl-10bOpen weightsStepFun · Step37.69Independentversion4.3Comparable-4.18obs. 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 →