AI model leaderboard

Every tracked model ranked across the headline benchmarks. The Intelligence Index averages each model's normalized scores; click any benchmark column header to sort by it.

Best overall

GPT 5.2 Thinking

90.8 Intelligence Index

Best at knowledge

GPT 5.4 Pro

94.4 GPQA Diamond

Best at math

Grok 4 Heavy

100.0 AIME 2025

Best at coding

Claude Opus 4.7

87.6 SWE-bench Verified

Best at multimodal

GPT 5 (Thinking)

84.2 MMMU

Models × benchmarks

Cells are best primary scores. Color intensity reflects normalized score. Click a column header to sort.

#	Model	MMLU-Pro	GPQA Diamond	Humanity's Last Exam	AIME 2025	SWE-bench Verified	LiveCodeBench	MMMU	AA-LCR	Intelligence Index
1	Deepseek V4 Pro	—	—	—	—	—	93.5%	—	—	—
2	Kimi K2.6	—	90.5%	54.0%	—	80.2%	89.6%	—	—	78.6 4/8
3	Kimi K2.5	—	—	—	—	76.8%	85.0%	—	—	—
4	Deepseek 3.2	85.0%	82.4%	40.8%	93.1%	73.1%	83.3%	—	—	76.3 6/8
5	GLM 4.6	—	81.0%	17.2%	93.9%	68.0%	82.8%	—	—	68.6 5/8
6	Gemma 4	85.2%	84.3%	—	—	—	80.0%	—	—	83.2 3/8
7	Grok 4 Heavy	—	88.4%	44.4%	100.0%	—	79.4%	—	—	78.1 4/8
8	Grok 3 Think	—	84.6%	—	93.3%	—	79.4%	—	—	85.8 3/8
9	Grok 4	—	87.5%	25.4%	91.7%	—	79.0%	—	—	70.9 4/8
10	DeepSeek V3.1 Terminus	85.0%	80.7%	21.7%	88.4%	—	74.9%	—	—	70.1 5/8
11	DeepSeek V3.2 Exp	85.0%	79.9%	—	89.3%	67.8%	74.1%	—	—	79.2 5/8
12	Qwen3-235B-A22B	—	—	—	81.5%	—	70.7%	—	—	—
13	Qwen3 235B A22B	—	—	—	81.5%	—	70.7%	—	—	—
14	Gemini 2.5 Pro	—	84.0%	18.8%	86.7%	63.8%	70.4%	—	—	64.7 5/8
15	Nemotron 3 Nano	78.3%	75.0%	—	89.1%	—	68.3%	—	—	77.7 4/8
16	Qwen3 30B A3B	—	65.8%	—	70.9%	—	62.6%	—	—	66.4 3/8
17	Grok 3	79.9%	75.4%	—	—	—	57.0%	73.2%	—	71.4 4/8
18	Kimi K2 Instruct	—	75.1%	—	49.5%	65.8%	53.7%	—	—	61.0 4/8
19	Magistral Medium	—	70.8%	—	64.9%	—	50.3%	—	—	62.0 3/8
20	Llama 4 Behemoth	82.2%	73.7%	—	—	—	49.4%	76.1%	—	70.4 4/8
21	Llama 4 Maverick	80.5%	69.8%	—	—	—	43.4%	73.4%	—	66.8 4/8
22	Grok 3 mini	78.9%	66.2%	—	—	—	41.5%	69.4%	—	64.0 4/8
23	Mistral Large 3	—	43.9%	—	—	—	34.4%	—	—	—
24	Gemini 2.5 Flash-Lite	—	64.6%	5.10%	49.8%	31.6%	33.7%	72.9%	—	43.0 6/8
25	Llama 4 Scout	74.3%	57.2%	—	—	—	32.8%	69.4%	—	58.4 4/8
26	Claude Haiku 3.5	41.6%	65.0%	—	—	40.6%	—	—	—	49.1 3/8
27	Claude Haiku 4.5	—	73.0%	—	80.7%	73.3%	—	73.2%	—	75.1 4/8
28	Claude Opus 3	—	50.4%	—	—	—	—	—	—	—
29	Claude Opus 4.5	—	87.0%	—	—	80.9%	—	80.7%	—	82.9 3/8
30	Claude Opus 4.6	—	91.3%	—	—	80.8%	—	—	—	—
31	Claude Opus 4.7	—	94.2%	46.9%	—	87.6%	—	—	—	76.2 3/8
32	Claude Sonnet 3.7	—	62.3%	—	—	62.3%	—	71.8%	—	65.5 3/8
33	Claude Sonnet 3.7 (Thinking)	—	78.2%	—	—	62.3%	—	75.0%	—	71.8 3/8
34	Claude Sonnet 4	—	75.4%	—	70.5%	72.7%	—	74.4%	—	73.3 4/8
35	Claude Sonnet 4.5	—	83.4%	—	87.0%	77.2%	—	77.8%	—	81.4 4/8
36	Claude Sonnet 4.6	—	89.9%	33.2%	—	79.6%	—	—	—	67.6 3/8
37	Command A	69.6%	50.8%	—	—	—	—	—	—	—
38	DeepSeek 3.2 Speciale	—	—	30.6%	96.0%	—	—	—	—	—
39	DeepSeek V3	75.9%	59.1%	—	—	42.0%	—	—	—	59.0 3/8
40	DeepSeek-R1	84.0%	71.5%	—	70.0%	49.2%	—	—	—	68.7 4/8
41	Devstral 2	—	—	—	—	72.2%	—	—	—	—
42	Gemini 2.5 Flash (Thinking)	—	82.8%	11.0%	72.0%	60.4%	—	—	—	56.6 4/8
43	Gemini 2.5 Pro (Thinking)	—	86.4%	21.6%	88.0%	59.6%	—	—	—	63.9 4/8
44	Gemini 3 Deep Think	—	93.8%	41.0%	—	—	—	—	—	—
45	Gemini 3 Flash	—	90.4%	—	—	78.0%	—	—	—	—
46	Gemini 3 Flash (Thinking)	—	90.4%	33.7%	95.2%	78.0%	—	—	—	74.3 4/8
47	Gemini 3 Pro	—	91.9%	37.5%	95.0%	76.2%	—	—	—	75.2 4/8
48	Gemini 3.1 Pro	—	94.3%	44.4%	—	80.6%	—	—	—	73.1 3/8
49	Gemma 3	78.0%	72.6%	—	—	—	—	—	—	—
50	GLM 5	—	86.0%	—	—	77.8%	—	—	—	—
51	GLM-5.1	—	86.2%	31.0%	—	—	—	—	—	—
52	GPT 4.1	—	66.3%	—	—	55.0%	—	75.0%	—	65.4 3/8
53	GPT 5	—	77.8%	6.30%	61.9%	52.8%	—	74.4%	—	54.6 5/8
54	GPT 5 (Thinking)	—	85.7%	24.8%	94.6%	74.9%	—	84.2%	—	72.8 5/8
55	GPT 5.1	—	88.1%	—	94.6%	74.9%	—	84.2%	—	85.5 4/8
56	GPT 5.1 Thinking	—	88.1%	—	94.6%	—	—	—	—	—
57	GPT 5.2 Pro	—	93.2%	—	—	—	—	—	—	—
58	GPT 5.2 Thinking	—	92.4%	—	100.0%	80.0%	—	—	—	90.8 3/8
59	GPT 5.3 Codex	—	92.6%	—	—	56.8%	—	—	—	—
60	GPT 5.4	—	92.8%	—	—	57.7%	—	—	—	—
61	GPT 5.4 Mini	—	88.0%	—	—	—	—	—	—	—
62	GPT 5.4 Nano	—	82.8%	—	—	—	—	—	—	—
63	GPT 5.4 Pro	—	94.4%	—	—	—	—	—	—	—
64	GPT 5.5	—	93.6%	41.4%	—	—	—	—	—	—
65	GPT 5.5 Instant	—	—	—	81.2%	—	—	—	—	—
66	GPT OSS 120B	90.0%	80.1%	—	—	—	—	—	—	—
67	GPT-4 Turbo	—	50.4%	—	—	—	—	—	—	—
68	GPT-4o	—	53.6%	—	—	—	—	69.1%	—	—
69	Grok Code Fast 1	—	—	—	—	70.8%	—	—	—	—
70	Llama 3.1 Nemotron Ultra	—	76.0%	—	—	—	—	—	—	—
71	Llama 3.2	—	32.8%	—	—	—	—	—	—	—
72	Llama 3.3	68.9%	50.5%	—	—	—	—	—	—	—
73	MiniMax M2.5	—	—	—	—	80.2%	—	—	—	—
74	Mistral Medium 3.5	—	—	—	—	77.6%	—	—	—	—
75	Mistral Small 3	66.3%	—	—	—	—	—	—	—	—
76	Muse Spark	—	89.5%	42.8%	—	77.4%	—	—	—	69.9 3/8
77	Nemotron 3 Super	75.7%	60.0%	—	—	—	—	—	—	—
78	Nova Lite	—	42.0%	—	—	—	—	—	—	—
79	Nova Micro	—	40.0%	—	—	—	—	—	—	—
80	Nova Premier	—	—	—	—	42.4%	—	—	—	—
81	Nova Pro	—	46.9%	—	—	—	—	—	—	—
82	o1	—	78.0%	8.12%	79.2%	48.9%	—	77.6%	—	58.4 5/8
83	o3	—	83.3%	20.3%	88.9%	69.1%	—	82.9%	—	68.9 5/8
84	o4 mini	—	—	—	92.7%	68.1%	—	81.6%	—	80.8 3/8
85	Opus 4.1 Thinking	—	80.9%	—	—	74.5%	—	—	—	—
86	Phi 4 reasoning plus	76.0%	69.3%	—	78.0%	—	—	—	—	74.4 3/8
87	Pixtral 12B	—	—	—	—	—	—	52.0%	—	—
88	Pixtral Large	—	—	—	—	—	—	64.0%	—	—
89	Qwen 3.5 122B A10B	—	86.6%	—	—	72.0%	—	76.9%	—	78.5 3/8
90	Qwen 3.5 27B	86.1%	85.5%	—	—	72.4%	—	—	—	81.3 3/8
91	Qwen 3.5 35B A3B	—	84.2%	—	—	69.2%	—	—	—	—
92	Qwen3 Coder	—	—	—	—	67.0%	—	—	—	—
93	Qwen3-30B-A3B	—	65.8%	—	—	—	—	—	—	—
94	R1 1776	—	71.5%	—	70.0%	—	—	—	—	—
95	Seed 1.5	80.1%	65.0%	—	—	—	—	73.9%	—	73.0 3/8

Capability scatter

Each dot is a model. Position shows two-axis capability; size reflects how many headline benchmarks the model has been scored on.

X axis Y axis

Go to section

Search

AI model leaderboard

Models × benchmarks

Capability scatter

Help

People also viewed

Create AI Tools

Mini Tool

Vibe code an AI Tool

Choose listing type: