Capability ranking

Best AI for Reasoning

Best Reasoning Models

Updated 2026-08-11

Ranked models99

Vendors17

Open-weight models17

Official prices40

Claude Opus 4.8 leads this page at 99.0, 0.0 points ahead of GPT-5.5.

Gemma 4 31B is the highest-ranked open-weight option at #7. Nova Micro has the lowest official input price among ranked models.

Top reasoning modelClaude Opus 4.899.0 reasoning score Best open-weight modelGemma 4 31BRank #7 Lowest official inputNova Micro$0.04 / 1M Largest contextGemini 1.5 Pro2.1M tokens

Top reasoning models

ModelReasoningKnowledgeMathCodingInstruction

Claude Opus 4.8Anthropic99.094.6N/A85.7N/A

GPT-5.5OpenAI99.083.762.574.3N/A

Qwen3.8 MaxAlibaba Cloud / Qwen Team97.670.7N/A80.1100.0

Gemini 3.1 ProGoogle96.290.3N/A54.0N/A

Qwen3.7 MaxAlibaba Cloud / Qwen Team91.589.193.185.692.2

Reasoning scoreOverall score

80 models with comparable dataOpen a model by selecting its logo

GPQA94.3%Gemini 3.1 Pro85 compared models

MMLU-Pro89.6%Qwen3.7 Max65 compared models

AIME 2025100.0%Gemini 3 Pro28 compared models

SWE-Bench Verified88.6%Claude Opus 4.826 compared models

MMLU90.6%Qwen3 VL 235B A22B Thinking55 compared models

Humanity's Last Exam58.4%Muse Spark30 compared models

LiveCodeBench79.0%Grok-413 compared models

MATH89.0%Gemma 3 27B35 compared models


01	ANClaude Opus 4.8Anthropic	Claude Opus 4.8	99.0	87.9	94.6	N/A	85.7	N/A	60.0%	1M	Closed weights	$5	$25
02	OPGPT-5.5OpenAI	GPT-5.5	99.0	84.0	83.7	62.5	74.3	N/A	80.0%	1.1M	Closed weights	$5	$30
03	ACQwen3.8 MaxAlibaba Cloud / Qwen Team	Qwen3.8 Max	97.6	89.3	70.7	N/A	80.1	100.0	80.0%	1M	Closed weights	$2	$6
04	GOGemini 3.1 ProGoogle	Gemini 3.1 Pro	96.2	77.5	90.3	N/A	54.0	N/A	60.0%	1M	Closed weights	$2	$12
05	ACQwen3.7 MaxAlibaba Cloud / Qwen Team	Qwen3.7 Max	91.5	78.8	89.1	93.1	85.6	92.2	100.0%	1M	Closed weights	$2.5	$7.5
06	OPGPT-5.4OpenAI	GPT-5.4	91.2	77.1	63.0	81.3	55.2	N/A	80.0%	1.1M	Closed weights	$2.5	$15
07	GOGemma 4 31BGoogle	Gemma 4 31B	87.6	60.1	67.4	35.3	61.5	N/A	80.0%	262.1K	Open weights	N/A	N/A
08	ACQwen3.7 PlusAlibaba Cloud / Qwen Team	Qwen3.7-Plus	84.7	73.8	83.5	66.9	61.6	87.8	100.0%	1M	Closed weights	$0.50	$3
09	XIMiMo V2.5 ProXiaomi	MiMo-V2.5-Pro	84.0	46.7	62.8	92.8	49.7	N/A	80.0%	1M	Open weights	$0.435	$0.87
10	ANClaude Opus 4.6Anthropic	Claude Opus 4.6	83.2	79.6	87.7	95.6	74.7	N/A	80.0%	1M	Closed weights	$5	$25
11	ACQwen3.5 397B A17BAlibaba Cloud / Qwen Team	Qwen3.5-397B-A17B	81.3	66.3	82.8	50.1	57.4	81.9	100.0%	262.1K	Open weights	$0.60	$3.6
12	ACQwen3.6 PlusAlibaba Cloud / Qwen Team	Qwen3.6 Plus	80.9	67.5	87.3	73.3	51.8	72.7	100.0%	1M	Closed weights	$0.50	$3
13	GOGemma 4 26B A4BGoogle	Gemma 4 26B-A4B	79.2	52.6	52.8	26.5	57.7	N/A	80.0%	262.1K	Open weights	N/A	N/A
14	OPGPT-5.2-ProOpenAI	GPT-5.2 Pro	78.5	71.6	59.8	99.1	N/A	N/A	60.0%	400K	Closed weights	$21	$168
15	ANClaude Sonnet 4.6Anthropic	Claude Sonnet 4.6	78.2	72.6	77.7	N/A	48.7	N/A	60.0%	1M	Closed weights	$3	$15
16	ANClaude Sonnet 3.5Anthropic	Claude 3.5 Sonnet	77.8	32.1	52.0	79.9	55.7	N/A	80.0%	200K	Closed weights	N/A	N/A
17	OPGPT-5.2OpenAI	GPT-5.2	74.0	71.4	69.9	90.0	87.5	N/A	80.0%	400K	Closed weights	$1.75	$14
18	ANClaude Opus 3Anthropic	Claude 3 Opus	73.8	16.7	30.9	55.4	46.1	N/A	80.0%	200K	Closed weights	N/A	N/A
19	GOGemini 3 ProGoogle	Gemini 3 Pro	73.4	67.3	74.5	98.2	54.4	N/A	80.0%	1M	Closed weights	N/A	N/A
20	GOGemini 3 FlashGoogle	Gemini 3 Flash	72.1	64.8	69.0	94.7	53.4	N/A	80.0%	1M	Closed weights	$0.50	$3
21	ACQwen3.6 27BAlibaba Cloud / Qwen Team	Qwen3.6-27B	71.8	60.5	71.4	49.9	49.0	N/A	80.0%	262.1K	Open weights	$0.60	$3.6
22	ACQwen3 235B A22B ThinkingAlibaba Cloud / Qwen Team	Qwen3-235B-A22B-Thinking-2507	71.4	48.5	61.9	63.5	53.9	84.9	100.0%	262.1K	Closed weights	N/A	N/A
23	COCommand R+Cohere	Command R+	71.2	0.0	30.8	10.6	N/A	N/A	60.0%	128K	Closed weights	$0.15	$0.60
24	GOGemma 4 12BGoogle	Gemma 4 12B	70.3	38.2	30.2	17.6	29.3	36.7	100.0%	N/A	Closed weights	N/A	N/A
25	AMNova ProAmazon	Nova Pro	68.6	20.9	65.2	69.9	75.4	86.7	100.0%	300K	Closed weights	$0.80	$3.2
26	GOGemma 2 27BGoogle	Gemma 2 27B	68.6	0.0	59.4	10.7	8.6	N/A	80.0%	N/A	Closed weights	N/A	N/A
27	MEMuse SparkMeta	Muse Spark	68.0	73.2	97.8	N/A	47.2	N/A	60.0%	N/A	Closed weights	N/A	N/A
28	OPGPT-4OpenAI	GPT-4	67.6	4.6	70.2	7.1	10.8	N/A	80.0%	8.2K	Closed weights	$30	$60
29	ACQwen3.5 122B A10BAlibaba Cloud / Qwen Team	Qwen3.5-122B-A10B	67.4	59.2	84.6	56.3	54.2	69.8	100.0%	262.1K	Open weights	$0.40	$3.2
30	MELlama 3.1 405BMeta	Llama 3.1 405B Instruct	67.2	24.5	41.4	75.0	75.4	62.5	100.0%	128K	Closed weights	N/A	N/A
31	ACQwen3.5 27BAlibaba Cloud / Qwen Team	Qwen3.5-27B	62.5	57.2	78.0	59.4	47.9	70.6	100.0%	262.1K	Open weights	$0.30	$2.4
32	GOGemini 1.5 ProGoogle	Gemini 1.5 Pro	62.0	20.5	46.1	66.9	41.5	75.7	100.0%	2.1M	Closed weights	N/A	N/A
33	ACQwen3.5 35B A3BAlibaba Cloud / Qwen Team	Qwen3.5-35B-A3B	61.8	52.0	71.2	43.8	44.2	54.4	100.0%	262.1K	Open weights	$0.25	$2
34	MIPhi 3.5 MoEMicrosoft	Phi-3.5-MoE-instruct	61.8	4.9	50.6	34.3	39.7	75.0	100.0%	N/A	Closed weights	N/A	N/A
35	ANClaude Opus 4.5Anthropic	Claude Opus 4.5	61.3	69.5	87.5	N/A	76.2	N/A	60.0%	200K	Closed weights	$5	$25
36	ACQwen3 235B A22BAlibaba Cloud / Qwen Team	Qwen3-235B-A22B-Instruct-2507	61.1	42.4	70.6	21.5	51.2	72.8	100.0%	262.1K	Open weights	$0.70	$2.8
37	GODiffusionGemma 26B A4BGoogle	DiffusionGemma 26B-A4B	59.9	34.2	37.9	11.8	21.2	N/A	80.0%	N/A	Closed weights	N/A	N/A
38	NRHermes 3 70BNous Research	Hermes 3 70B	59.2	7.2	39.6	1.4	N/A	82.8	80.0%	131.1K	Closed weights	N/A	N/A
39	ACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team	Qwen3 VL 235B A22B Thinking	58.6	45.8	55.9	53.8	46.1	77.7	100.0%	262.1K	Closed weights	N/A	N/A
40	ANClaude Sonnet 3Anthropic	Claude 3 Sonnet	58.4	3.5	18.0	35.7	20.0	N/A	80.0%	200K	Closed weights	N/A	N/A
41	ACQwen3 Next 80B A3B ThinkingAlibaba Cloud / Qwen Team	Qwen3-Next-80B-A3B-Thinking	58.2	40.9	65.2	45.7	38.5	62.7	100.0%	131.1K	Open weights	$0.50	$6
42	NVLlama 3.1 Nemotron 70BNVIDIA	Llama 3.1 Nemotron 70B Instruct	57.3	3.1	52.0	53.2	N/A	0.0	80.0%	128K	Open weights	N/A	N/A
43	GOGemini 1.5 FlashGoogle	Gemini 1.5 Flash	55.1	10.7	28.1	52.4	23.1	37.7	100.0%	1M	Closed weights	N/A	N/A
44	ACQwen3 VL 235B A22BAlibaba Cloud / Qwen Team	Qwen3 VL 235B A22B Instruct	54.0	43.2	64.8	28.0	53.2	76.8	100.0%	262.1K	Closed weights	N/A	N/A
45	OPGPT-4-TurboOpenAI	GPT-4 Turbo	53.8	14.6	71.7	54.3	55.4	N/A	80.0%	128K	Closed weights	$10	$30
46	ACQwen3 Next 80B A3BAlibaba Cloud / Qwen Team	Qwen3-Next-80B-A3B-Instruct	53.4	36.6	53.0	18.9	49.0	62.7	100.0%	131.1K	Open weights	$0.50	$2
47	MELlama 3.1 70BMeta	Llama 3.1 70B Instruct	52.9	13.2	24.2	100.0	32.3	51.6	100.0%	128K	Closed weights	N/A	N/A
48	XAGrok 4xAI	Grok-4	51.9	56.5	64.1	72.3	83.3	N/A	80.0%	256K	Closed weights	N/A	N/A
49	AMNova LiteAmazon	Nova Lite	51.4	13.2	41.9	63.7	48.5	71.9	100.0%	300K	Closed weights	$0.06	$0.24
50	ACQwen2 72BAlibaba Cloud / Qwen Team	Qwen2 72B Instruct	50.9	9.8	24.9	41.4	46.6	N/A	80.0%	N/A	Closed weights	N/A	N/A
51	ACQwen2.5 32BAlibaba Cloud / Qwen Team	Qwen2.5 32B Instruct	50.6	14.8	37.4	81.7	68.3	N/A	80.0%	N/A	Closed weights	$0.70	$2.8
52	GOGemma 2 9BGoogle	Gemma 2 9B	49.9	0.0	39.5	5.3	2.3	N/A	80.0%	N/A	Closed weights	N/A	N/A
53	MELongCat Flash ChatMeituan	LongCat-Flash-Chat	49.3	36.9	75.0	38.2	49.6	70.3	100.0%	128K	Closed weights	N/A	N/A
54	ACQwen3 VL 32B ThinkingAlibaba Cloud / Qwen Team	Qwen3 VL 32B Thinking	47.9	40.5	66.5	45.1	33.6	63.4	100.0%	N/A	Closed weights	N/A	N/A
55	ACQwen3.5 9BAlibaba Cloud / Qwen Team	Qwen3.5-9B	47.6	41.4	52.0	28.1	33.6	44.6	100.0%	262.1K	Open weights	N/A	N/A
56	OPo3OpenAI	o3	47.3	52.6	22.3	37.3	64.5	67.9	100.0%	200K	Closed weights	$2	$8
57	MAMistral NeMoMistral AI	Mistral NeMo Instruct	46.9	0.0	17.8	N/A	N/A	N/A	40.0%	128K	Closed weights	$0.15	$0.15
58	GOGemma 3 27BGoogle	Gemma 3 27B	46.8	15.8	37.9	89.0	39.2	61.1	100.0%	131.1K	Closed weights	N/A	N/A
59	ANClaude Haiku 3Anthropic	Claude 3 Haiku	46.4	0.0	24.8	23.1	27.7	N/A	80.0%	200K	Closed weights	N/A	N/A
60	ACQwen2.5 Coder 32BAlibaba Cloud / Qwen Team	Qwen2.5-Coder 32B Instruct	46.3	4.4	17.2	39.3	68.0	N/A	80.0%	128K	Closed weights	$0.287	$0.861
61	ALJamba 1.5 LargeAI21 Labs	Jamba 1.5 Large	46.1	2.7	33.9	31.9	N/A	14.3	80.0%	256K	Closed weights	N/A	N/A
62	GOGemma 4 E4BGoogle	Gemma 4 E4B	45.2	24.4	27.1	5.9	18.3	N/A	80.0%	131.1K	Open weights	N/A	N/A
63	ANClaude Haiku 3.5Anthropic	Claude 3.5 Haiku	42.2	9.3	22.7	42.9	35.5	N/A	80.0%	200K	Closed weights	N/A	N/A
64	AMNova MicroAmazon	Nova Micro	40.1	4.3	29.3	50.7	34.6	47.7	100.0%	128K	Closed weights	$0.035	$0.14
65	GOGemma 3 12BGoogle	Gemma 3 12B	39.4	11.2	35.4	74.8	32.4	51.4	100.0%	131.1K	Closed weights	N/A	N/A
66	ACQwen3 VL 32BAlibaba Cloud / Qwen Team	Qwen3 VL 32B Instruct	39.3	35.2	43.3	19.5	9.6	41.0	100.0%	N/A	Closed weights	N/A	N/A
67	ACQwen3 VL 30B A3B ThinkingAlibaba Cloud / Qwen Team	Qwen3 VL 30B A3B Thinking	37.8	31.8	43.4	38.8	30.8	39.9	100.0%	262.1K	Closed weights	N/A	N/A
68	ANClaude Opus 4Anthropic	Claude Opus 4	37.7	46.4	68.8	31.9	57.2	N/A	80.0%	200K	Closed weights	N/A	N/A
69	ACQwen3.5 4BAlibaba Cloud / Qwen Team	Qwen3.5-4B	35.8	32.4	36.9	18.8	25.0	36.3	100.0%	N/A	Closed weights	N/A	N/A
70	OPGPT-4o-miniOpenAI	GPT-4o mini	35.3	8.1	50.5	46.4	28.5	N/A	80.0%	128K	Closed weights	$0.15	$0.60
71	ALJamba 1.5 MiniAI21 Labs	Jamba 1.5 Mini	34.8	0.0	11.2	14.9	N/A	0.0	80.0%	256.1K	Closed weights	N/A	N/A
72	MAMinistral 8BMistral AI	Ministral 8B Instruct	34.1	0.0	12.9	27.1	0.0	27.3	100.0%	131.1K	Open weights	N/A	N/A
73	MIPhi 4Microsoft	Phi 4	33.8	9.4	20.6	74.3	38.5	3.9	100.0%	16K	Closed weights	$0.125	$0.50
74	GOGemma 4 E2BGoogle	Gemma 4 E2B	32.9	11.6	14.3	0.0	11.5	N/A	80.0%	131.1K	Open weights	N/A	N/A
75	ACQwen3 VL 8B ThinkingAlibaba Cloud / Qwen Team	Qwen3 VL 8B Thinking	32.5	28.5	47.0	33.6	28.9	41.3	100.0%	262.1K	Closed weights	N/A	N/A
76	ACQwen3 VL 30B A3BAlibaba Cloud / Qwen Team	Qwen3 VL 30B A3B Instruct	30.6	28.8	41.6	14.3	7.7	37.6	100.0%	262.1K	Closed weights	N/A	N/A
77	ACQwen3 VL 4B ThinkingAlibaba Cloud / Qwen Team	Qwen3 VL 4B Thinking	28.4	23.0	30.2	20.9	13.5	29.2	100.0%	262.1K	Closed weights	N/A	N/A
78	MIPhi 3.5 miniMicrosoft	Phi-3.5-mini-instruct	27.7	0.0	42.9	24.4	18.7	100.0	100.0%	128K	Closed weights	N/A	N/A
79	IBGranite 3.3 8BIBM	Granite 3.3 8B Instruct	27.4	3.8	48.7	25.3	80.8	10.2	100.0%	128K	Closed weights	N/A	N/A
80	MELlama 3.1 8BMeta	Llama 3.1 8B Instruct	26.4	0.0	10.2	0.0	18.5	20.3	100.0%	131.1K	Closed weights	N/A	N/A
81	GOGemma 3n E4B LiteRTGoogle	Gemma 3n E4B Instructed LiteRT Preview	25.7	0.0	26.4	19.9	16.9	29.6	100.0%	N/A	Closed weights	N/A	N/A
82	MIPhi 4 MiniMicrosoft	Phi 4 Mini	23.8	0.0	48.2	35.6	N/A	N/A	60.0%	128K	Open weights	$0.075	$0.30
83	MELlama 3.2 3BMeta	Llama 3.2 3B Instruct	23.2	0.0	4.0	16.7	N/A	14.1	80.0%	128K	Closed weights	N/A	N/A
84	ACQwen2.5 Coder 7BAlibaba Cloud / Qwen Team	Qwen2.5-Coder 7B Instruct	22.8	0.0	3.7	19.9	50.5	N/A	80.0%	N/A	Closed weights	$0.144	$0.287
85	OPGPT-4oOpenAI	GPT-4o	22.8	23.9	35.2	0.0	8.5	17.4	100.0%	128K	Closed weights	$2.5	$10
86	ACQwen3 VL 8BAlibaba Cloud / Qwen Team	Qwen3 VL 8B Instruct	22.4	24.8	25.4	5.2	3.9	36.0	100.0%	262.1K	Closed weights	N/A	N/A
87	GOGemma 3 4BGoogle	Gemma 3 4B	21.7	0.0	13.3	51.8	10.8	50.4	100.0%	131.1K	Closed weights	N/A	N/A
88	ACQwen2.5 14BAlibaba Cloud / Qwen Team	Qwen2.5 14B Instruct	20.3	8.2	40.0	73.1	51.2	N/A	80.0%	N/A	Closed weights	$0.35	$1.4
89	OPGPT-3.5-TurboOpenAI	GPT-3.5 Turbo	18.1	0.0	17.2	10.7	12.3	N/A	80.0%	16.4K	Closed weights	$0.50	$1.5
90	GOGemini DiffusionGoogle	Gemini Diffusion	16.3	3.6	46.1	3.5	36.8	N/A	80.0%	N/A	Closed weights	N/A	N/A
91	IBIBM Granite 4.0 TinyIBM	IBM Granite 4.0 Tiny Preview	15.9	0.0	25.1	8.5	36.9	3.9	100.0%	N/A	Closed weights	N/A	N/A
92	ACQwen3.5 2BAlibaba Cloud / Qwen Team	Qwen3.5-2B	15.1	2.8	13.3	N/A	N/A	11.3	60.0%	N/A	Closed weights	N/A	N/A
93	ACQwen3 VL 4BAlibaba Cloud / Qwen Team	Qwen3 VL 4B Instruct	12.3	19.3	33.4	4.0	1.9	16.8	100.0%	262.1K	Closed weights	N/A	N/A
94	GOGemma 3n E2B LiteRTGoogle	Gemma 3n E2B Instructed LiteRT (Preview)	9.5	0.0	10.3	6.5	7.0	11.4	100.0%	N/A	Closed weights	N/A	N/A
95	GOGemma 3n E4BGoogle	Gemma 3n E4B Instructed	8.0	0.0	23.4	19.9	16.9	29.6	100.0%	32K	Closed weights	N/A	N/A
96	BAERNIE 4.5Baidu	ERNIE 4.5	6.5	0.0	0.3	0.0	0.0	0.0	100.0%	128K	Closed weights	N/A	N/A
97	ACQwen3.5 0.8BAlibaba Cloud / Qwen Team	Qwen3.5-0.8B	2.2	0.0	3.3	N/A	N/A	0.6	60.0%	N/A	Closed weights	N/A	N/A
98	GOGemma 3n E2BGoogle	Gemma 3n E2B Instructed	1.8	0.0	11.0	6.5	7.0	11.4	100.0%	N/A	Closed weights	N/A	N/A
99	GOGemma 3 1BGoogle	Gemma 3 1B	0.1	0.0	0.6	6.9	1.0	11.7	100.0%	N/A	Closed weights	N/A	N/A

Official PAYG prices appear here. Third-party offers remain on the pricing and model detail pages.

Reasoning scoreOfficial API price blend (8:1 input/output), USD / 1M

Efficient frontier40 models with official PAYG prices

Top-model vendor mix

25 models

Alibaba Cloud / Qwen Team7

Access model

Open weights

Closed weights

Official price available

Claude Opus 4.8vsGPT-5.5 Claude Opus 4.8vsGemma 4 31B Claude Opus 4.8vsNova Micro

Common questions about Best AI for Reasoning.

What is the best model on Best AI for Reasoning?

Claude Opus 4.8 is currently ranked first with a reasoning score of 99.0.

How is this leaderboard ranked?

Each model appears once using its current scored version. Category pages use the capability named in the title, while the overall page uses the LLMBoard score.

How many models are included?

This page currently ranks 99 unique model products.

Where does pricing come from?

Main price columns use the model vendor's official standard PAYG API rate. Eligible third-party offers appear only in separately labeled columns, and unavailable official prices display as N/A.

Does Arena affect the score?

No. Arena results are displayed as an independent signal and are not included in the current LLMBoard capability score.