AA Intelligence Index

Artificial Analysis' composite of its own evaluations (GPQA, HLE, IFBench, LCR, τ²-Bench, Terminal-Bench Hard, SciCode, AA-Omniscience and more).

Published by

Artificial Analysis

Category

Composite

Index weight

1.0

Models

539

Data as of

9 Sept 2026

Source: Artificial Analysis (artificialanalysis.ai). Data taken from the public leaderboard.

1. 1Claude Fable 5.153.4
2. 2Claude Fable 5.153.2
3. 3GPT-6 Astra52.8
4. 4GPT-6 Astra52.5
5. 5Claude Fable 5.151.2
6. 6GPT-6 Astra51.0
7. 7Claude Opus 550.7
8. 8Claude Fable 549.7
9. 9GPT-6 Astra49.7
10. 10Claude Opus 549.6
11. 11Claude Fable 5.149.1
12. 12Claude Opus 548.2
13. 13Muse Spark 1.348.2
14. 14GPT-5.6 Sol47.1
15. 15Claude Fable 5.147.0

SearchAll providersAI21 LabsAI9StarsAlibabaAllen Institute for AIAllenaiAmazonAnt GroupAnt-groupAnthropicApodexArcee AIBaiduByteDanceCelerisChina MobileCohereDeepSeekGoogleIBMInceptionInclusionAIKimiKorea TelecomKwaiKATLG AI ResearchLiquid AILongCatMBZUAI Institute of Foundation ModelsMetaMicrosoftMiniMaxMistral AIMoonshot AIMotif TechnologiesMultiverse ComputingNanbeigeNaverNex AGINous ResearchNVIDIAOpenAIOpenBMBOpenChat / BerkeleyPerplexityPrime IntellectQwenRekaReka AISapiens AISarvamServiceNowSK TelecomStepFunSwiss AI InitiativeThinking MachinesTII UAETrillion LabsUnknownUpstagexAIXiaomiZhipu AIOpen weights539 of 539

#	Model	Score ↓	Run ↕	Index ↕	Published ↕
1	Claude Fable 5.1Anthropic	53.4	Claude Fable 5.1 (max with fallback)	70.0	–
2	Claude Fable 5.1Anthropic	53.2	–	70.3	–
3	GPT-6 AstraOpenAI	52.8	GPT-6 Astra (max)	69.9	–
4	GPT-6 AstraOpenAI	52.5	–	68.6	–
5	Claude Fable 5.1Anthropic	51.2	–	69.7	–
6	GPT-6 AstraOpenAI	51.0	–	69.5	–
7	Claude Opus 5Anthropic	50.7	Claude Opus 5 (max)	70.0	–
8	Claude Fable 5Anthropic	49.7	Claude Fable 5 (with fallback)	70.4	–
9	GPT-6 AstraOpenAI	49.7	–	67.4	–
10	Claude Opus 5Anthropic	49.6	–	66.3	–
11	Claude Fable 5.1Anthropic	49.1	–	68.0	–
12	Claude Opus 5Anthropic	48.2	–	69.3	–
13	Muse Spark 1.3Meta	48.2	Muse Spark 1.3 (max)	–	–
14	GPT-5.6 SolOpenAI	47.1	GPT-5.6 Sol (max)	65.3	–
15	Claude Fable 5.1Anthropic	47.0	–	65.9	–
16	GPT-6 AstraOpenAI	46.0	–	65.8	–
17	GPT-6 AstraOpenAI	45.2	–	64.4	–
18	Muse Spark 1.3Meta	45.2	–	63.9	–
19	Claude Opus 5Anthropic	45.1	–	63.1	–
20	GLM-5.3Zhipu AIopen	44.9	GLM-5.3 (max)	–	–
21	Grok 4.6xAI	44.4	Grok 4.6 (high)	63.6	–
22	Grok 4.6xAI	44.3	–	62.3	–
23	GPT-5.6 SolOpenAI	44.1	–	67.5	–
24	Kimi K3Moonshot AIopen	43.8	Kimi K3 (max)	66.5	–
25	Grok 4.6xAI	43.0	–	63.8	–
26	GPT-5.6 SolOpenAI	42.5	–	68.0	–
27	GPT-5.6 TerraOpenAI	42.3	GPT-5.6 Terra (max)	62.3	–
28	Qwen3.8-Flash-NextAlibabaopen	42.2	–	59.3	–
29	Claude Opus 4.8Anthropic	42.0	Claude Opus 4.8 (max)	64.7	–
30	GLM-5.3-FlashZhipu AIopen	41.9	–	61.0	–
31	Gemini 3.8 FlashGoogle	41.2	Gemini 3.8 Flash (high)	62.4	–
32	Claude Opus 4.7Anthropic	40.7	Claude Opus 4.7 (max)	65.5	–
33	Qwen3 8Alibaba	40.3	–	63.9	–
34	Qwen3.8 2.4T A95BAlibabaopen	40.0	–	–	–
35	Gemini 3.8 FlashGoogle	40	–	63.5	–
36	Muse Spark 1.2Meta	39.8	Muse Spark 1.2 (xhigh)	62.6	–
37	Claude Opus 5Anthropic	39.8	–	61.7	–
38	Gemini 3.7 FlashGoogle	39.6	–	64.4	–
39	GPT-5.6 SolOpenAI	39.5	–	66.0	–
40	Gemini 3.7 FlashGoogle	39.4	Gemini 3.7 Flash (high)	61.7	–
41	Grok 4.5xAI	39.1	Grok 4.5 (high)	63.4	–
42	GPT-5.4OpenAI	39.0	GPT-5.4 (xhigh)	63.4	–
43	GLM-5.2Zhipu AIopen	38.6	GLM-5.2 (max)	57.4	–
44	GPT-5.5OpenAI	38.6	GPT-5.5 (xhigh)	66.6	–
45	Claude Sonnet 5Anthropic	38.4	Claude Sonnet 5 (max)	61.2	–
46	GPT-5.6 TerraOpenAI	38.2	–	64.2	–
47	GPT-5.6 LunaOpenAI	37.5	GPT-5.6 Luna (max)	56.6	–
48	GPT-5.5OpenAI	37.3	–	66.3	–
49	Gemini 3.7 FlashGoogle	37.0	–	62.5	–
50	DeepSeek V4 ProDeepSeekopen	36.3	DeepSeek V4 Pro 0813 (max)	59.7	–
51	Grok 4.6xAI	35.4	–	59.6	–
52	SAgnes 2.5 Pro BetaSapiens AI	35.2	–	–	–
53	Deepseek v4 Flash VisionDeepSeek	35.0	DeepSeek V4 Flash Vision (max)	57.4	–
54	GPT-5.6 LunaOpenAI	34.8	–	59.3	–
55	DeepSeek V4 FlashDeepSeekopen	34.5	DeepSeek V4 Flash 0731 (max)	59.1	–
56	GPT-5.6 TerraOpenAI	34.5	–	63.1	–
57	Kimi K3Moonshot AIopen	34.5	–	58.8	–
58	Gemini 3.6 FlashGoogle	34.3	Gemini 3.6 Flash	60.7	–
59	Muse Spark 1.1Meta	34.3	Muse Spark 1.1 (xhigh)	64.9	–
60	GPT-5.5OpenAI	34.1	–	64.9	–

Show 100 more