AISOTA· State of the Art

Daily AI Decision Intelligence

Multi-source rankings blended into one score. See which model is truly State of the Art — today.

Data as of 2026-08-25Updated Aug 25, 11:16 AM

Overall Intelligence

Multi-source blend of general intelligence benchmarks into a single AISOTA score.

#	Model	Artificial Analysis	LiveBench	AISOTA Score
1SOTA	Claude Fable 5 (with fallback)2 sources	90.0	97.9	94.0
22	gpt-5.5-xhigh	—	93.6	93.6
33	Claude Opus 5 (max)2 sources	95.0	91.5	93.3
4	GPT-5.6 Sol (max)2 sources	85.0	95.7	90.3
5	smaug-agentic	—	87.2	87.2
6	qwen3.8-max	—	85.1	85.1
7	Grok 4.6 (high)2 sources	80.0	80.9	80.5
8	Kimi K3 (max)2 sources	75.0	83.0	79.0
9	gpt-5.4-xhigh	—	76.6	76.6
10	gemini-3.1-pro-preview-high	—	70.2	70.2

Multi-source blend of coding benchmarks into a single AISOTA score.

#	Model	Artificial Analysis Agentic	LiveBench Coding	AISOTA Score
1SOTA	smaug-agentic	—	97.9	97.9
22	Claude Opus 5 (max)2 sources	95.0	93.6	94.3
33	GLM-5.3 (max)2 sources	90.0	89.4	89.7
4	qwen3.8-max	—	87.2	87.2
5	claude-sonnet-5-xhigh-effort	—	85.1	85.1
6	Claude Fable 5 (with fallback)2 sources	70.0	95.7	82.8
7	GPT-5.6 Sol (max)2 sources	80.0	83.0	81.5
8	Kimi K3 (max)2 sources	65.0	91.5	78.3
9	deepseek-v4-flash-vision-exp	—	76.6	76.6
10	Grok 4.6 (high)2 sources	85.0	68.1	76.5

Intelligence per dollar, based on median input + output price.

#	Model	AA Intelligence	OpenRouter Price	AISOTA Score
1SOTA	Gemini 3.7 Flash (high)2 sources	50.0	65.9	58.0
22	GPT-5.6 Luna (max)2 sources	40.0	74.2	57.1
33	GPT-5.6 Sol (max)2 sources	85.0	22.7	53.9
4	Claude Opus 5 (max)2 sources	95.0	11.0	53.0
5	Grok 4.6 (high)2 sources	80.0	19.1	49.5
6	Claude Fable 5 (with fallback)2 sources	90.0	5.7	47.9
7	GLM-5.3 (max)2 sources	70.0	24.1	47.0
8	Muse Spark 1.2 (xhigh)2 sources	60.0	26.1	43.0
9	Qwen3.8 2.4T A95B2 sources	65.0	19.4	42.2
10	Kimi K3 (max)2 sources	75.0	7.7	41.4

API request volume per model over the last 7 days, sourced from OpenRouter public rankings.

#	Model	OpenRouter Usage	AISOTA Score
1SOTA	deepseek-v4-flash	953.94M	99.8
22	gpt-5-6-luna	288.9M	99.5
33	gemini-2-5-flash-lite	261.27M	99.3
4	ox-alpha	216.08M	99.1
5	gpt-oss-120b	132.51M	98.8
6	gpt-4o-mini	127.04M	98.6
7	gemini-3-1-flash-lite	116.4M	98.4
8	mimo-v2-5	116.1M	98.1
9	gemini-2-5-flash	113.98M	97.9
10	text-embedding-3-small	112.52M	97.7