Benchmark Leaderboard

MMLU

120 models with MMLU scores, ranked from highest to lowest.

Score Leaderboard

#	Model	Developer	Score
1	Google: Gemini 2.5 Pro	google	89.5
2	Google: Gemini 2.5 Flash	google	88.2
3	MiniMax: MiniMax M2.5	minimax	87.5
4	Anthropic: Claude 3.7 Sonnet (thinking)	anthropic	87.5
5	OpenAI: GPT-5	openai	87.1
6	OpenAI: GPT-5 Codex	openai	86.5
7	DeepSeek: DeepSeek V3.2 Speciale	deepseek	86.3
8	OpenAI: GPT-5.1-Codex	openai	86.0
9	OpenAI: GPT-5.2-Codex	openai	86.0
10	OpenAI: o3	openai	85.3
11	DeepSeek: R1	deepseek	84.9
12	MoonshotAI: Kimi K2 Thinking	moonshotai	84.8
13	Qwen: Qwen3 235B A22B Thinking 2507	qwen	84.3
14	Qwen: Qwen3 Max	qwen	84.1
15	OpenAI: o1	openai	84.1
16	DeepSeek: DeepSeek V3.2	deepseek	83.7
17	Google: Gemini 2.5 Pro Preview 05-06	google	83.7
18	OpenAI: GPT-5 Mini	openai	83.7
19	Qwen: Qwen3 VL 235B A22B Thinking	qwen	83.6
20	DeepSeek: DeepSeek V3.1 Terminus	deepseek	83.6
21	OpenAI: o4 Mini	openai	83.2
22	Nous: Hermes 3 405B Instruct	nousresearch	82.9
23	Qwen: Qwen3 235B A22B Instruct 2507	qwen	82.8
24	xAI: Grok 3 Mini	x-ai	82.8
25	MoonshotAI: Kimi K2.5	moonshotai	82.4
26	Qwen: Qwen3 Max Thinking	qwen	82.4
27	Qwen: Qwen3 Next 80B A3B Thinking	qwen	82.4
28	Qwen: Qwen3 VL 235B A22B Instruct	qwen	82.3
29	Prime Intellect: INTELLECT-3	prime-intellect	82.2
30	MiniMax: MiniMax M2	minimax	82.0
31	OpenAI: GPT-5.1-Codex-Mini	openai	82.0
32	MoonshotAI: Kimi K2 0905	moonshotai	81.9
33	Qwen: Qwen3 Next 80B A3B Instruct	qwen	81.9
34	Qwen: Qwen3 VL 32B Instruct	qwen	81.8
35	Z.ai: GLM 4.5 Air	z-ai	81.5
36	NVIDIA: Llama 3.3 Nemotron Super 49B V1.5	nvidia	81.4
37	Kwaipilot: KAT-Coder-Pro V1	kwaipilot	81.3
38	OpenAI: gpt-oss-120b	openai	80.8
39	MiniMax: MiniMax M1	minimax	80.8
40	Mistral Large	mistralai	80.7
41	Qwen: Qwen3 VL 30B A3B Thinking	qwen	80.7
42	Qwen: Qwen3 30B A3B Thinking 2507	qwen	80.5
43	Upstage: Solar Pro 3	upstage	80.5
44	OpenAI: o3 Mini High	openai	80.2
45	Z.ai: GLM 4.6V	z-ai	79.9
46	Google: Gemini 2.5 Flash Lite Preview 09-2025	google	79.6
47	DeepSeek: R1 Distill Llama 70B	deepseek	79.5
48	NVIDIA: Nemotron 3 Nano 30B A3B (free)	nvidia	79.4
49	xAI: Grok Code Fast 1	x-ai	79.3
50	OpenAI: o3 Mini	openai	79.1
51	OpenAI: GPT-5 Nano	openai	78.0
52	Qwen: Qwen3 30B A3B	qwen	77.7
53	Baidu: ERNIE 4.5 300B A47B	baidu	77.6
54	OpenAI: GPT-4.1 Mini	openai	77.5
55	Anthropic: Claude Sonnet 4.6	anthropic	77.2
56	Qwen: Qwen3 VL 30B A3B Instruct	qwen	76.4
57	AllenAI: Olmo 3.1 32B Think	allenai	76.3
58	Qwen: Qwen3 235B A22B	qwen	76.2
59	Anthropic: Claude Haiku 4.5	anthropic	76.0
60	NVIDIA: Nemotron Nano 12B 2 VL (free)	nvidia	75.9
61	Google: Gemini 2.5 Flash Lite	google	75.9
62	Perplexity: Sonar Pro	perplexity	75.5
63	Z.ai: GLM 4.5V	z-ai	75.1
64	Qwen: Qwen3 VL 8B Thinking	qwen	74.9
65	OpenAI: GPT-4o	openai	74.8
66	Xiaomi: MiMo-V2-Flash	xiaomi	74.4
67	xAI: Grok 4 Fast	x-ai	74.3
68	xAI: Grok 4.1 Fast	x-ai	74.3
69	NVIDIA: Nemotron Nano 9B V2 (free)	nvidia	74.2
70	NVIDIA: Nemotron Nano 9B V2	nvidia	74.2
71	OpenAI: GPT-4o (2024-05-13)	openai	74.0
72	DeepSeek: R1 Distill Qwen 32B	deepseek	73.9
73	Meta: Llama 3.1 405B Instruct	meta-llama	73.2
74	Qwen: Qwen3 32B	qwen	72.7
75	Qwen: Qwen3 30B A3B Instruct 2507	qwen	72.5
76	Google: Gemini 2.0 Flash Lite	google	72.4
77	Qwen: Qwen2.5 VL 72B Instruct	qwen	72.0
78	OpenAI: gpt-oss-20b	openai	71.8
79	NVIDIA: Llama 3.1 Nemotron 70B Instruct	nvidia	71.3
80	Meta: Llama 3.3 70B Instruct	meta-llama	71.3
81	Cohere: Command A	cohere	71.2
82	Mistral: Devstral Medium	mistralai	70.8
83	Qwen: Qwen3 Coder 30B A3B Instruct	qwen	70.6
84	xAI: Grok 3 Beta	x-ai	70.3
85	Mistral: Pixtral Large 2411	mistralai	70.1
86	Anthropic: Claude Opus 4.5	anthropic	69.6
87	Anthropic: Claude Opus 4.6	anthropic	69.6
88	OpenAI: GPT-4 Turbo	openai	69.4
89	Perplexity: Sonar	perplexity	68.9
90	Qwen: Qwen3 VL 8B Instruct	qwen	68.6
91	Mistral: Mistral Medium 3.1	mistralai	68.3
92	Mistral: Mistral Medium 3	mistralai	68.3
93	Mistral Large 2407	mistralai	68.3
94	Mistral: Mistral Small 3	mistralai	68.1
95	Mistral: Devstral Small 1.1	mistralai	67.8
96	Qwen: Qwen3 14B	qwen	67.5
97	Nous: Hermes 3 70B Instruct	nousresearch	66.4
98	AllenAI: Olmo 3 7B Think	allenai	65.5
99	NVIDIA: Nemotron Nano 12B 2 VL	nvidia	64.9
100	Qwen: QwQ 32B	qwen	64.8
101	OpenAI: GPT-4o-mini	openai	64.8
102	Qwen: Qwen3 8B	qwen	64.3
103	Qwen: Qwen2.5 VL 32B Instruct	qwen	63.5
104	Qwen2.5 Coder 32B Instruct	qwen	63.5
105	Qwen: Qwen-Turbo	qwen	63.3
106	Mistral: Saba	mistralai	61.1
107	NVIDIA: Nemotron 3 Nano 30B A3B	nvidia	57.9
108	AI21: Jamba Large 1.7	ai21	57.7
109	AllenAI: Olmo 3 7B Instruct	allenai	52.2
110	LiquidAI: LFM2-8B-A1B	liquid	50.5
111	Google: Gemma 3n 4B	google	48.8
112	Meta: Llama 3.1 8B Instruct	meta-llama	47.6
113	Qwen: Qwen2.5 Coder 7B Instruct	qwen	47.3
114	Meta: Llama 3.2 11B Vision Instruct	meta-llama	46.4
115	OpenAI: GPT-3.5 Turbo	openai	46.2
116	Cohere: Command R+ (08-2024)	cohere	43.2
117	Mistral: Mixtral 8x7B Instruct	mistralai	38.7
118	Meta: Llama 3.2 3B Instruct	meta-llama	34.7
119	Mistral: Mistral 7B Instruct v0.1	mistralai	24.5
120	Meta: Llama 3.2 1B Instruct	meta-llama	20.0

Value Rankings

Best performance per dollar — score divided by avg price per million tokens.

#	Model	Score	Avg $/M	Value
1	LiquidAI: LFM2-8B-A1B	50.5	$0.01	3366.7
2	Google: Gemma 3n 4B	48.8	$0.03	1626.7
3	Meta: Llama 3.1 8B Instruct	47.6	$0.03	1360.0
4	Mistral: Mistral Small 3	68.1	$0.07	1047.7
5	Qwen: Qwen3 235B A22B Instruct 2507	82.8	$0.09	968.4
6	Meta: Llama 3.2 11B Vision Instruct	46.4	$0.05	946.9
7	OpenAI: gpt-oss-20b	71.8	$0.09	844.7
8	Qwen: Qwen2.5 Coder 7B Instruct	47.3	$0.06	788.3
9	Qwen: Qwen-Turbo	63.3	$0.08	779.1
10	NVIDIA: Nemotron Nano 9B V2	74.2	$0.10	742.0
11	OpenAI: gpt-oss-120b	80.8	$0.11	705.7
12	NVIDIA: Nemotron 3 Nano 30B A3B	57.9	$0.12	463.2
13	Qwen: Qwen3 32B	72.7	$0.16	454.4
14	Qwen: Qwen3 14B	67.5	$0.15	450.0
15	Qwen: Qwen3 30B A3B	77.7	$0.18	431.7
16	Qwen: Qwen3 Coder 30B A3B Instruct	70.6	$0.17	415.3
17	Qwen: Qwen3 30B A3B Thinking 2507	80.5	$0.20	411.8
18	AllenAI: Olmo 3 7B Think	65.5	$0.16	409.4
19	Xiaomi: MiMo-V2-Flash	74.4	$0.19	391.6
20	Google: Gemini 2.0 Flash Lite	72.4	$0.19	386.1
21	Qwen: Qwen3 30B A3B Instruct 2507	72.5	$0.20	371.8
22	AllenAI: Olmo 3 7B Instruct	52.2	$0.15	348.0
23	OpenAI: GPT-5 Nano	78.0	$0.22	346.7
24	Meta: Llama 3.3 70B Instruct	71.3	$0.21	339.5
25	Mistral: Devstral Small 1.1	67.8	$0.20	339.0
26	NVIDIA: Llama 3.3 Nemotron Super 49B V1.5	81.4	$0.25	325.6
27	Google: Gemini 2.5 Flash Lite Preview 09-2025	79.6	$0.25	318.4
28	Qwen2.5 Coder 32B Instruct	63.5	$0.20	317.5
29	Qwen: Qwen3 VL 32B Instruct	81.8	$0.26	314.6
30	Google: Gemini 2.5 Flash Lite	75.9	$0.25	303.6
31	Qwen: Qwen3 8B	64.3	$0.22	285.8
32	DeepSeek: DeepSeek V3.2	83.7	$0.32	257.5
33	DeepSeek: R1 Distill Qwen 32B	73.9	$0.29	254.8
34	Qwen: Qwen3 235B A22B Thinking 2507	84.3	$0.35	237.5
35	Qwen: Qwen3 VL 8B Instruct	68.6	$0.29	236.6
36	Qwen: QwQ 32B	64.8	$0.27	235.6
37	Qwen: Qwen3 VL 30B A3B Instruct	76.4	$0.33	235.1
38	AllenAI: Olmo 3.1 32B Think	76.3	$0.33	234.8
39	Nous: Hermes 3 70B Instruct	66.4	$0.30	221.3
40	Upstage: Solar Pro 3	80.5	$0.38	214.7
41	xAI: Grok 4 Fast	74.3	$0.35	212.3
42	xAI: Grok 4.1 Fast	74.3	$0.35	212.3
43	xAI: Grok 3 Mini	82.8	$0.40	207.0
44	Meta: Llama 3.2 3B Instruct	34.7	$0.20	177.5
45	Meta: Llama 3.2 1B Instruct	20.0	$0.11	176.2
46	OpenAI: GPT-4o-mini	64.8	$0.38	172.8
47	DeepSeek: DeepSeek V3.1 Terminus	83.6	$0.50	167.2
48	Z.ai: GLM 4.5 Air	81.5	$0.49	166.3
49	Mistral: Mistral 7B Instruct v0.1	24.5	$0.15	163.3
50	NVIDIA: Nemotron Nano 12B 2 VL	64.9	$0.40	162.3
51	Qwen: Qwen2.5 VL 32B Instruct	63.5	$0.40	158.8
52	Kwaipilot: KAT-Coder-Pro V1	81.3	$0.52	157.1
53	Mistral: Saba	61.1	$0.40	152.8
54	Qwen: Qwen3 VL 235B A22B Instruct	82.3	$0.54	152.4
55	Qwen: Qwen3 Next 80B A3B Instruct	81.9	$0.60	137.6
56	Z.ai: GLM 4.6V	79.9	$0.60	133.2
57	MiniMax: MiniMax M2	82.0	$0.63	130.7
58	Prime Intellect: INTELLECT-3	82.2	$0.65	126.5
59	Qwen: Qwen3 Next 80B A3B Thinking	82.4	$0.67	122.1
60	MiniMax: MiniMax M2.5	87.5	$0.75	117.1
61	Baidu: ERNIE 4.5 300B A47B	77.6	$0.69	112.5
62	DeepSeek: DeepSeek V3.2 Speciale	86.3	$0.80	107.9
63	DeepSeek: R1 Distill Llama 70B	79.5	$0.75	106.0
64	Qwen: Qwen3 VL 8B Thinking	74.9	$0.74	101.1
65	xAI: Grok Code Fast 1	79.3	$0.85	93.3
66	Qwen: Qwen2.5 VL 72B Instruct	72.0	$0.80	90.0
67	Nous: Hermes 3 405B Instruct	82.9	$1.00	82.9
68	OpenAI: GPT-4.1 Mini	77.5	$1.00	77.5
69	OpenAI: GPT-5 Mini	83.7	$1.13	74.4
70	OpenAI: GPT-5.1-Codex-Mini	82.0	$1.13	72.9
71	Mistral: Mixtral 8x7B Instruct	38.7	$0.54	71.7
72	Perplexity: Sonar	68.9	$1.00	68.9
73	MoonshotAI: Kimi K2 Thinking	84.8	$1.23	68.7
74	MoonshotAI: Kimi K2 0905	81.9	$1.20	68.3
75	Qwen: Qwen3 235B A22B	76.2	$1.14	67.0
76	Google: Gemini 2.5 Flash	88.2	$1.40	63.0
77	Z.ai: GLM 4.5V	75.1	$1.20	62.6
78	MoonshotAI: Kimi K2.5	82.4	$1.33	62.2
79	MiniMax: MiniMax M1	80.8	$1.30	62.2
80	NVIDIA: Llama 3.1 Nemotron 70B Instruct	71.3	$1.20	59.4
81	Mistral: Devstral Medium	70.8	$1.20	59.0
82	Mistral: Mistral Medium 3.1	68.3	$1.20	56.9
83	Mistral: Mistral Medium 3	68.3	$1.20	56.9
84	DeepSeek: R1	84.9	$1.60	53.1
85	OpenAI: GPT-3.5 Turbo	46.2	$1.00	46.2
86	Qwen: Qwen3 Max Thinking	82.4	$2.34	35.2
87	OpenAI: o4 Mini	83.2	$2.75	30.3
88	OpenAI: o3 Mini High	80.2	$2.75	29.2
89	OpenAI: o3 Mini	79.1	$2.75	28.8
90	Anthropic: Claude Haiku 4.5	76.0	$3.00	25.3
91	Qwen: Qwen3 Max	84.1	$3.60	23.4
92	Mistral Large	80.7	$4.00	20.2
93	Meta: Llama 3.1 405B Instruct	73.2	$4.00	18.3
94	Mistral: Pixtral Large 2411	70.1	$4.00	17.5
95	Mistral Large 2407	68.3	$4.00	17.1
96	OpenAI: o3	85.3	$5.00	17.1
97	Google: Gemini 2.5 Pro	89.5	$5.63	15.9
98	OpenAI: GPT-5	87.1	$5.63	15.5
99	OpenAI: GPT-5 Codex	86.5	$5.63	15.4
100	OpenAI: GPT-5.1-Codex	86.0	$5.63	15.3
101	Google: Gemini 2.5 Pro Preview 05-06	83.7	$5.63	14.9
102	OpenAI: GPT-4o	74.8	$6.25	12.0
103	AI21: Jamba Large 1.7	57.7	$5.00	11.5
104	Cohere: Command A	71.2	$6.25	11.4
105	OpenAI: GPT-5.2-Codex	86.0	$7.88	10.9
106	Anthropic: Claude 3.7 Sonnet (thinking)	87.5	$9.00	9.7
107	Anthropic: Claude Sonnet 4.6	77.2	$9.00	8.6
108	Perplexity: Sonar Pro	75.5	$9.00	8.4
109	xAI: Grok 3 Beta	70.3	$9.00	7.8
110	OpenAI: GPT-4o (2024-05-13)	74.0	$10.00	7.4
111	Cohere: Command R+ (08-2024)	43.2	$6.25	6.9
112	Anthropic: Claude Opus 4.5	69.6	$15.00	4.6
113	Anthropic: Claude Opus 4.6	69.6	$15.00	4.6
114	OpenAI: GPT-4 Turbo	69.4	$20.00	3.5
115	OpenAI: o1	84.1	$37.50	2.2