vals Gpqa Diamond

126 models evaluated

#	Model	Provider	Type	Score
1	Google	google/gemini-3.1-pro-preview	Gemini 3.1 Pro Preview	95.45%
2	OpenAI	openai/gpt-5.6-sol	GPT-5.6 Sol	95.20%
3	Google	google/gemini-3.6-flash	Gemini 3.6 Flash	93.43%
4	Claude Opus 5	Anthropic	93.43%
5	Claude Fable 5	Anthropic	93.18%
6	GPT-5.5	OpenAI	93.18%
7	Kimi K3	Moonshot AI	92.93%
8	Grok 4.5	xAI	92.93%
9	Gemini 3.5 Flash	Google	92.68%
10	MiniMax M3	MiniMax	92.68%
11	Claude Opus 4.8	Anthropic	92.42%
12	GPT-5.6 Luna	OpenAI	91.67%
13	Gemini 3 Pro Preview	Google	91.67%
14	GPT-5.2	OpenAI	91.67%
15	GPT-5.4	OpenAI	91.67%
16	Grok 4.3	xAI	91.41%
17	Muse Spark 1.1	Meta	91.16%
18	GPT-5.6 Terra	OpenAI	90.91%
19	Qwen3.7 Max	Alibaba	90.15%
20	Claude Opus 4.7	Anthropic	90.15%
21	Muse Spark	Meta	89.65%
22	Claude Opus 4.6 Thinking	Anthropic	89.65%
23	DeepSeek V4 Pro	DeepSeek	89.39%
24	Kimi K2.6	Moonshot AI	89.14%
25	Claude Sonnet 5	Anthropic	88.89%
26	Grok 4.20 0309 Reasoning	xAI	88.64%
27	Grok 4 0709	xAI	88.13%
28	Gemini 3 Flash Preview	Google	87.88%
29	Qwen3.5 Plus Thinking	Alibaba	87.37%
30	Qwen3.6 Plus	Alibaba	87.37%
31	Inkling	Thinkingmachines	87.12%
32	GPT-5.1	OpenAI	86.62%
33	MiniMax M2.7	MiniMax	86.62%
34	Nemotron 3 Ultra 550b A55b	Nvidia	86.11%
35	Claude Opus 4.5 20251101 Thinking	Anthropic	85.86%
36	GPT-5	OpenAI	85.61%
37	GLM 5.2	Zhipu AI	85.61%
38	Claude Sonnet 4.6	Anthropic	85.61%
39	Grok 4 Fast Reasoning	xAI	85.35%
40	Qwen3 Max	Alibaba	84.85%
41	GLM 5.1	Zhipu AI	84.52%
42	Grok 4.1 Fast Reasoning	xAI	84.34%
43	O3	OpenAI	84.09%
44	Kimi K2.5 Thinking	Moonshot AI	84.09%
45	Gemini 3.5 Flash Lite	Google	83.84%
46	GLM 5 Thinking	Zhipu AI	83.33%
47	GPT-5.4 Mini	OpenAI	83.08%
48	Qwen3.5 Flash	Alibaba	82.83%
49	Mimo V2.5 Pro	Xiaomi	82.58%
50	MiniMax M2.5	MiniMax	82.07%
51	Claude Sonnet 4.5 20250929 Thinking	Anthropic	81.63%
52	Gemini 2.5 Flash Preview 09 2025	Google	81.57%
53	Mimo V2.5	Xiaomi	81.57%
54	Gemini 3.1 Flash Lite Preview	Google	81.06%
55	Gemini 2.5 Pro Exp 03 25	Google	80.81%
56	GPT-5 Mini	OpenAI	80.30%
57	DeepSeek V3p2 Thinking	Fireworks AI	80.30%
58	GLM 4.7	Zhipu AI	80.05%
59	Claude Opus 4.5	Anthropic	79.55%
60	Qwen3 Max	Alibaba	79.55%
61	Grok 3 Mini Fast High Reasoning	xAI	79.29%
62	GPT Oss 120b	Fireworks AI	78.54%
63	MiniMax M2.1	MiniMax	78.54%
64	Kimi K2 Thinking	Moonshot AI	78.54%
65	Qwen3 Max Preview	Alibaba	77.78%
66	GPT-5.4 Nano	OpenAI	77.53%
67	Gemini 2.5 Flash Preview 09 2025 Thinking	Google	76.52%
68	Claude Opus 4.1 20250805 Thinking	Anthropic	76.26%
69	DeepSeek V3p2	Fireworks AI	76.26%
70	O3 Mini	OpenAI	75.50%
71	Claude 3.7 Sonnet 20250219 Thinking	Anthropic	75.25%
72	Claude Sonnet 4 20250514 Thinking	Anthropic	75.00%
73	O4 Mini	OpenAI	74.50%
74	GLM 4.6	Zhipu AI	74.50%
75	Grok 3	xAI	74.24%
76	O1	OpenAI	73.23%
77	Grok 3 Mini Fast Low Reasoning	xAI	72.98%
78	Claude Haiku 4.5 20251001 Thinking	Anthropic	72.22%
79	GLM 4.5	Zhipu AI	72.22%
80	Claude Opus 4	Anthropic	71.72%
81	Moonshotai Kimi K2 Instruct	Together AI	71.46%
82	Gemini 2.5 Flash Lite Preview 09 2025 Thinking	Google	70.20%
83	Qwen3 235b A22b	Fireworks AI	70.20%
84	Claude Opus 4.1	Anthropic	69.95%
85	Llama4 Maverick Instruct Basic	Fireworks AI	69.44%
86	Claude Sonnet 4	Anthropic	69.44%
87	GPT Oss 20b	Fireworks AI	68.94%
88	Mistral Large 2512	Mistral AI	68.43%
89	GPT-4.1 Mini	OpenAI	67.93%
90	Claude 3.7 Sonnet	Anthropic	67.42%
91	GPT-4.1	OpenAI	65.40%
92	Gemini 2.0 Flash 001	Google	65.15%
93	Grok 4.1 Fast Non Reasoning	xAI	65.15%
94	Gemini 2.5 Flash Lite Preview 09 2025	Google	64.14%
95	GPT-5 Nano	OpenAI	63.38%
96	Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561 Thinking	Together AI	62.37%
97	Magistral Medium 2509	Mistral AI	62.37%
98	Grok 4 Fast Non Reasoning	xAI	62.12%
99	DeepSeek V3 0324	Fireworks AI	61.62%
100	Claude 3.5 Sonnet	Anthropic	59.34%
101	Mimo V2 Flash	Xiaomi	59.34%
102	Gemini 1.5 Pro 002	Google	58.33%
103	Magistral Small 2509	Mistral AI	58.33%
104	Laguna Xs.2	Poolside	58.08%
105	Gemini 2.5 Flash Preview 04 17	Google	57.58%
106	DeepSeek V3	Fireworks AI	54.55%
107	GPT-4o	OpenAI	53.79%
108	GPT-4.1 Nano	OpenAI	50.76%
109	Mistral Small 2402	Mistral AI	50.76%
110	Grok 2 1212	xAI	50.76%
111	GPT-4o	OpenAI	50.25%
112	Meta Llama Llama 3.3 70B Instruct Turbo	Together AI	50.00%
113	Command A 03 2025	Cohere	48.48%
114	Mistral Large 2411	Mistral AI	47.73%
115	Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561	Together AI	47.73%
116	Meta Llama Llama 4 Scout 17B 16E Instruct	Together AI	46.97%
117	Gemini 1.5 Flash 002	Google	45.96%
118	Mistral Small 2503	Mistral AI	44.19%
119	GPT-4o Mini	OpenAI	44.19%
120	Claude 3.5 Haiku	Anthropic	37.88%
121	Jamba Large 1.6	AI21 Labs	36.11%
122	Mistral Medium 3.5	Mistral AI	34.85%
123	Jamba Mini 1.6	AI21 Labs	32.07%
124	Command R Plus	Cohere	31.06%
125	GPT-3.5 Turbo	OpenAI	30.56%
126	Laguna M.1	Poolside	27.02%