BenchmarkActivecategory · compositefamily · artificial-analysis-intelligence-index · variant index

Artificial Analysis Intelligence Index

artificialanalysis.ai

composite of several evaluations run by Artificial Analysis

data quality57

Updated 7 h ago · first seen 11 Sept 2026

Metric: index ↑
Current results: 1,272
Models: 477
Current leader: Claude Fable 5.1 53.4

Score history · Claude Opus 5 10 rows

Claude Opus 5

49.65aa_slug=claude-opus-5-xhigh · version=4.3 · estimated=false · reasoning_effort=xhigh12 Sept 2026
39.79aa_slug=claude-opus-5-low · version=4.3 · estimated=false · reasoning_effort=low12 Sept 2026
48.24aa_slug=claude-opus-5-high · version=4.3 · estimated=false · reasoning_effort=high12 Sept 2026
50.7aa_slug=claude-opus-5 · version=4.3 · estimated=false · reasoning_effort=max12 Sept 2026
45.06aa_slug=claude-opus-5-medium · version=4.3 · estimated=false · reasoning_effort=medium12 Sept 2026
49.65aa_slug=claude-opus-5-xhigh · version=4.3 · estimated=false · aa_variant_slug=claude-opus-5-xhigh11 Sept 2026
39.79aa_slug=claude-opus-5-low · version=4.3 · estimated=false · aa_variant_slug=claude-opus-5-low11 Sept 2026
48.24aa_slug=claude-opus-5-high · version=4.3 · estimated=false · aa_variant_slug=claude-opus-5-high11 Sept 2026
50.7aa_slug=claude-opus-5 · version=4.3 · estimated=false11 Sept 2026
45.06aa_slug=claude-opus-5-medium · version=4.3 · estimated=false · aa_variant_slug=claude-opus-5-medium11 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.

53.4Claude Fable 5.1 Anthropic Independent11 Sept 2026
53.2Claude Fable 5.1 Anthropic Independent11 Sept 2026
51.2Claude Fable 5.1 Anthropic Independent11 Sept 2026
51.0gpt-6-astra OpenAI Independent11 Sept 2026
49.7gpt-6-astra OpenAI Independent11 Sept 2026
45.1Claude Opus 5 Anthropic Independent11 Sept 2026
14.6grok-3-mini-reasoning SpaceXAI Independent11 Sept 2026
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 · trust independent-evaluator

Select models with +, then Compare.

Leaderboard
#	Model	Score	Trust	Configuration	vs leader	Evaluated	Source	Actions
401	Mixtral 8x22BOpen weightsMistral AI · Mixtral 8 · best of 2 rows	5.74	Independent	reasoningoffversion4.3	Partially comparable0.00	obs. 12 Sept 2026	artificialanalysis.aiT2	History
402	llama-2-chat-7bOpen weightsMeta AI · Llama 2 · best of 2 rows	5.71	Independent	reasoningoffversion4.3	Partially comparable-0.03	obs. 12 Sept 2026	artificialanalysis.aiT2	History
403	Llama-3.2-3BRestricted weightsMeta AI · Llama 3.2 · best of 2 rows	5.70	Independent	reasoningoffversion4.3	Partially comparable-0.04	obs. 12 Sept 2026	artificialanalysis.aiT2	History
404	minicpm-v4-6-1-3bOpen weightsOpenBMB · best of 2 rows	5.69	Independent	reasoningoffversion4.3	Partially comparable-0.05	obs. 12 Sept 2026	artificialanalysis.aiT2	History
405	jamba-reasoning-3bOpen weightsAI21 Labs · Jamba · best of 2 rows	5.67	Independent	reasoningonversion4.3	Partially comparable-0.07	obs. 12 Sept 2026	artificialanalysis.aiT2	History
406	Reka Flash 3Open weightsrekaai · best of 2 rows	5.65	Independent	reasoningonversion4.3	Partially comparable-0.09	obs. 12 Sept 2026	artificialanalysis.aiT2	History
406	qwen1.5-110b-chatOpen weightsAlibaba Group · Qwen1.5 · best of 2 rows	5.65	Independent	reasoningoffversion4.3	Partially comparable-0.09	obs. 12 Sept 2026	artificialanalysis.aiT2	History
408	Olmo-3-7B-ThinkOpen weightsAllen Institute for AI · OLMo 3 · best of 2 rows	5.62	Independent	reasoningonversion4.3	Partially comparable-0.12	obs. 12 Sept 2026	artificialanalysis.aiT2	History
409	claude-21ClosedAnthropic · Claude 21 · best of 2 rows	5.59	Independent	reasoningoffversion4.3	Partially comparable-0.15	obs. 12 Sept 2026	artificialanalysis.aiT2	History
410	Claude 3 HaikuClosedAnthropic · Claude · best of 2 rows	5.58	Independent	reasoningoffversion4.3	Partially comparable-0.16	obs. 12 Sept 2026	artificialanalysis.aiT2	History
410	olmo-2-7bOpen weightsAllen Institute for AI · OLMo 2 · best of 2 rows	5.58	Independent	reasoningoffversion4.3	Partially comparable-0.16	obs. 12 Sept 2026	artificialanalysis.aiT2	History
412	molmo-7b-dOpen weightsAllen Institute for AI · Molmo · best of 2 rows	5.56	Independent	reasoningoffversion4.3	Partially comparable-0.18	obs. 12 Sept 2026	artificialanalysis.aiT2	History
413	ling-mini-2-0Open weightsinclusionAI · best of 2 rows	5.54	Independent	reasoningoffversion4.3	Partially comparable-0.20	obs. 12 Sept 2026	artificialanalysis.aiT2	History
414	DeepSeek-R1-Distill-Qwen-1.5BOpen weightsDeepSeek · Qwen · best of 2 rows	5.51	Independent	reasoningonversion4.3	Partially comparable-0.23	obs. 12 Sept 2026	artificialanalysis.aiT2	History
414	claude-2ClosedAnthropic · Claude 2 · best of 2 rows	5.51	Independent	reasoningoffversion4.3	Partially comparable-0.23	obs. 12 Sept 2026	artificialanalysis.aiT2	History
414	deepseek-v2Open weightsDeepSeek · DeepSeek · best of 2 rows	5.51	Independent	reasoningoffversion4.3	Partially comparable-0.23	obs. 12 Sept 2026	artificialanalysis.aiT2	History
417	Mistral Small 1.0ClosedMistral AI · Mistral · best of 2 rows	5.50	Independent	reasoningoffversion4.3	Partially comparable-0.24	obs. 12 Sept 2026	artificialanalysis.aiT2	History
418	gpt-3.5-turboClosedOpenAI · GPT 3.5 · best of 2 rows	5.49	Independent	reasoningoffversion4.3	Partially comparable-0.25	obs. 12 Sept 2026	artificialanalysis.aiT2	History
418	mistral-mediumClosedMistral AI · Mistral · best of 2 rows	5.49	Independent	reasoningoffversion4.3	Partially comparable-0.25	obs. 12 Sept 2026	artificialanalysis.aiT2	History
420	Ministral 3 8BOpen weightsMistral AI · Ministral 3 · best of 2 rows	5.48	Independent	reasoningoffversion4.3	Partially comparable-0.26	obs. 12 Sept 2026	artificialanalysis.aiT2	History
421	llama-3-instruct-70bOpen weightsMeta AI · Llama 3 · best of 2 rows	5.45	Independent	reasoningoffversion4.3	Partially comparable-0.29	obs. 12 Sept 2026	artificialanalysis.aiT2	History
422	arctic-instructOpen weightsSnowflake · best of 2 rows	5.44	Independent	reasoningoffversion4.3	Partially comparable-0.30	obs. 12 Sept 2026	artificialanalysis.aiT2	History
422	qwen-chat-72bOpen weightsAlibaba Group · Qwen · best of 2 rows	5.44	Independent	reasoningoffversion4.3	Partially comparable-0.30	obs. 12 Sept 2026	artificialanalysis.aiT2	History
424	lfm-40bClosedLiquid AI · LFM · best of 2 rows	5.42	Independent	reasoningoffversion4.3	Partially comparable-0.32	obs. 12 Sept 2026	artificialanalysis.aiT2	History
425	llama-3-2-instruct-11b-visionOpen weightsMeta AI · Llama 3.2 · best of 2 rows	5.41	Independent	reasoningoffversion4.3	Partially comparable-0.33	obs. 12 Sept 2026	artificialanalysis.aiT2	History
426	palm-2ClosedGoogle · PaLM 2 · best of 2 rows	5.37	Independent	reasoningoffversion4.3	Partially comparable-0.37	obs. 12 Sept 2026	artificialanalysis.aiT2	History
427	deepseek-coder-v2-liteOpen weightsDeepSeek · DeepSeek · best of 2 rows	5.34	Independent	reasoningoffversion4.3	Partially comparable-0.40	obs. 12 Sept 2026	artificialanalysis.aiT2	History
427	gemini-1-0-proClosedGoogle · Gemini 1.0 · best of 2 rows	5.34	Independent	reasoningoffversion4.3	Partially comparable-0.40	obs. 12 Sept 2026	artificialanalysis.aiT2	History
429	sarvam-m-reasoningOpen weightsSarvam · best of 2 rows	5.31	Independent	reasoningonversion4.3	Partially comparable-0.43	obs. 12 Sept 2026	artificialanalysis.aiT2	History
430	command-r-plus-04-2024Open weightsCohere · Command · best of 2 rows	5.30	Independent	reasoningoffversion4.3	Partially comparable-0.44	obs. 12 Sept 2026	artificialanalysis.aiT2	History
430	deepseek-llm-67b-chatOpen weightsDeepSeek · DeepSeek · best of 2 rows	5.30	Independent	reasoningoffversion4.3	Partially comparable-0.44	obs. 12 Sept 2026	artificialanalysis.aiT2	History
430	llama-2-chat-13bOpen weightsMeta AI · Llama 2 · best of 2 rows	5.30	Independent	reasoningoffversion4.3	Partially comparable-0.44	obs. 12 Sept 2026	artificialanalysis.aiT2	History
430	llama-2-chat-70bOpen weightsMeta AI · Llama 2 · best of 2 rows	5.30	Independent	reasoningoffversion4.3	Partially comparable-0.44	obs. 12 Sept 2026	artificialanalysis.aiT2	History
434	dbrxOpen weightsDatabricks · best of 2 rows	5.29	Independent	reasoningoffversion4.3	Partially comparable-0.45	obs. 12 Sept 2026	artificialanalysis.aiT2	History
434	openchat-35Open weightsOpenChat · best of 2 rows	5.29	Independent	reasoningoffversion4.3	Partially comparable-0.45	obs. 12 Sept 2026	artificialanalysis.aiT2	History
436	exaone-4-0-1-2bOpen weightsLG AI Research · EXAONE 4.0 · best of 4 rows	5.27	Independent	reasoningonversion4.3	Partially comparable-0.47	obs. 12 Sept 2026	artificialanalysis.aiT2	History
437	Olmo-3-7B-InstructOpen weightsAllen Institute for AI · OLMo 3 · best of 2 rows	5.24	Independent	reasoningoffversion4.3	Partially comparable-0.50	obs. 12 Sept 2026	artificialanalysis.aiT2	History
438	jamba-1-7-miniOpen weightsAI21 Labs · Jamba 1.7 · best of 2 rows	5.22	Independent	reasoningoffversion4.3	Partially comparable-0.52	obs. 12 Sept 2026	artificialanalysis.aiT2	History
438	lfm2-5-1-2b-thinkingOpen weightsLiquid AI · LFM2.5 · best of 2 rows	5.22	Independent	reasoningonversion4.3	Partially comparable-0.52	obs. 12 Sept 2026	artificialanalysis.aiT2	History
440	LFM2.5-1.2B-InstructRestricted weightsLiquid AI · LFM2.5 · best of 2 rows	5.21	Independent	reasoningoffversion4.3	Partially comparable-0.53	obs. 12 Sept 2026	artificialanalysis.aiT2	History
440	jamba-1-5-miniOpen weightsAI21 Labs · Jamba 1.5 · best of 2 rows	5.21	Independent	reasoningoffversion4.3	Partially comparable-0.53	obs. 12 Sept 2026	artificialanalysis.aiT2	History
440	lfm2-2-6bOpen weightsLiquid AI · LFM2.2 · best of 2 rows	5.21	Independent	reasoningoffversion4.3	Partially comparable-0.53	obs. 12 Sept 2026	artificialanalysis.aiT2	History
443	granite-4-0-h-nano-1bOpen weightsIBM · Granite 4.0 · best of 2 rows	5.19	Independent	reasoningoffversion4.3	Partially comparable-0.55	obs. 12 Sept 2026	artificialanalysis.aiT2	History
443	qwen3-1.7b-instructOpen weightsAlibaba Group · Qwen3.1 · best of 4 rows	5.19	Independent	reasoningonversion4.3	Partially comparable-0.55	obs. 12 Sept 2026	artificialanalysis.aiT2	History
445	Qwen3 8BOpen weightsQwen · Qwen3	5.16	Independent	version4.3	Partially comparable-0.58	obs. 11 Sept 2026	artificialanalysis.aiT2	History
445	jamba-1-6-miniOpen weightsAI21 Labs · Jamba 1.6 · best of 2 rows	5.16	Independent	reasoningoffversion4.3	Partially comparable-0.58	obs. 12 Sept 2026	artificialanalysis.aiT2	History
447	Mixtral 8x7BOpen weightsMistral AI · Mixtral 8 · best of 2 rows	5.12	Independent	reasoningoffversion4.3	Partially comparable-0.62	obs. 12 Sept 2026	artificialanalysis.aiT2	History
447	gemma-3-270mOpen weightsGoogle · Gemma 3 · best of 2 rows	5.12	Independent	reasoningoffversion4.3	Partially comparable-0.62	obs. 12 Sept 2026	artificialanalysis.aiT2	History
449	Granite 4.0 MicroOpen weightsIBM · Granite 4.0 · best of 2 rows	5.11	Independent	reasoningoffversion4.3	Partially comparable-0.63	obs. 12 Sept 2026	artificialanalysis.aiT2	History
449	apertus-70b-instructOpen weightsSwiss AI Initiative · best of 2 rows	5.11	Independent	reasoningoffversion4.3	Partially comparable-0.63	obs. 12 Sept 2026	artificialanalysis.aiT2	History
451	deephermes-3-llama-3-1-8b-previewOpen weightsNous Research · Llama 3.1 · best of 2 rows	5.08	Independent	reasoningoffversion4.3	Partially comparable-0.66	obs. 12 Sept 2026	artificialanalysis.aiT2	History
452	Mistral 7BOpen weightsMistral AI · Mistral · best of 2 rows	5.03	Independent	reasoningoffversion4.3	Partially comparable-0.71	obs. 12 Sept 2026	artificialanalysis.aiT2	History
452	claude-instantClosedAnthropic · Claude · best of 2 rows	5.03	Independent	reasoningoffversion4.3	Partially comparable-0.71	obs. 12 Sept 2026	artificialanalysis.aiT2	History
452	command-r-03-2024Open weightsCohere · Command · best of 2 rows	5.03	Independent	reasoningoffversion4.3	Partially comparable-0.71	obs. 12 Sept 2026	artificialanalysis.aiT2	History
452	llama-65bOpen weightsMeta AI · Llama · best of 2 rows	5.03	Independent	reasoningoffversion4.3	Partially comparable-0.71	obs. 12 Sept 2026	artificialanalysis.aiT2	History
452	qwen-chat-14bOpen weightsAlibaba Group · Qwen · best of 2 rows	5.03	Independent	reasoningoffversion4.3	Partially comparable-0.71	obs. 12 Sept 2026	artificialanalysis.aiT2	History
457	granite-4-0-nano-1bOpen weightsIBM · Granite 4.0 · best of 2 rows	5.01	Independent	reasoningoffversion4.3	Partially comparable-0.73	obs. 12 Sept 2026	artificialanalysis.aiT2	History
458	Molmo2-8BOpen weightsAllen Institute for AI · best of 2 rows	5	Independent	reasoningoffversion4.3	Partially comparable-0.74	obs. 12 Sept 2026	artificialanalysis.aiT2	History
459	lfm2-8b-a1bOpen weightsLiquid AI · LFM2 · best of 2 rows	4.93	Independent	reasoningoffversion4.3	Partially comparable-0.81	obs. 12 Sept 2026	artificialanalysis.aiT2	History
460	granite-3-3-8b-instructOpen weightsIBM · Granite 3.3 · best of 2 rows	4.92	Independent	reasoningoffversion4.3	Partially comparable-0.82	obs. 12 Sept 2026	artificialanalysis.aiT2	History
461	Gemma 3 27BOpen weightsGoogle · Gemma 3 · best of 2 rows	4.85	Independent	reasoningoffversion4.3	Partially comparable-0.89	obs. 12 Sept 2026	artificialanalysis.aiT2	History
462	Ministral 3 3BOpen weightsMistral AI · Ministral 3 · best of 2 rows	4.84	Independent	reasoningoffversion4.3	Partially comparable-0.90	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	Gemma 3 4BRestricted weightsGoogle · Gemma 3 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	LFM2-1.2BRestricted weightsLiquid AI · LFM2.1 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	LFM2.5-VL-1.6BRestricted weightsLiquid AI · LFM2.5 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	Llama-3.2-1BRestricted weightsMeta AI · Llama 3.2 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	Qwen3-0.6BOpen weightsQwen · Qwen3.0	4.83	Independent	version4.3	Partially comparable-0.91	obs. 11 Sept 2026	artificialanalysis.aiT2	History
463	apertus-8b-instructOpen weightsSwiss AI Initiative · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	gemma-3-1bOpen weightsGoogle · Gemma 3 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	gemma-3n-e2bOpen weightsGoogle · Gemma 3 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	gemma-3n-e4bOpen weightsGoogle · Gemma 3 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	granite-4-0-350mOpen weightsIBM · Granite 4.0 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	granite-4-0-h-350mOpen weightsIBM · Granite 4.0 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	llama-3-instruct-8bOpen weightsMeta AI · Llama 3 · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	qwen3-0.6b-instructOpen weightsAlibaba Group · Qwen3.0 · best of 3 rows	4.83	Independent	reasoningonversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
463	tiny-aya-globalRestricted weightsCohere · Aya · best of 2 rows	4.83	Independent	reasoningoffversion4.3	Partially comparable-0.91	obs. 12 Sept 2026	artificialanalysis.aiT2	History
477	Gemma 3 12BOpen weightsGoogle · Gemma 3 · best of 2 rows	3.75	Independent	reasoningoffversion4.3	Partially comparable-1.99	obs. 12 Sept 2026	artificialanalysis.aiT2	History

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 →

Definition

Task: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 17 h agomedium
Metric: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 17 h agomedium
Metric range: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 7 h agomedium
Variant: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 7 h agomedium
Harness: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 7 h agomedium
Comparability note: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 7 h agomedium
Known limitations: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 17 h agomedium
Website: artificialanalysis.ai
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 17 h agomedium

Active group

index

477 models · 1,272 current rows

Comparability

Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric

Also known as

AA Intelligence Index, Artificial Analysis Index, Artificial Analysis Intelligence Index, Intelligence Index

slug artificial-analysis-intelligence-index · bench_01M293SPG0EAFKRT01VFHFK9SC

Rows are grouped by comparability group = canonical metric × config_key (hash of the task-defining configuration keys: variant, evaluator, harness, shots, pass regime…). A leaderboard shows one row per canonical model: its best current row inside the group. Reasoning effort, temperature or judge differences keep rows in the same group but mark them partially comparable. delta_rank compares with the ranking built from the closed (previous) rows of the same group.