vals Live Code Bench

131 models evaluated

#	Model	Provider	Type	Score
1	Anthropic	anthropic/claude-fable-5	Claude Fable 5	89.78%
2	Anthropic	anthropic/claude-opus-5	Claude Opus 5	89.03%
3	Google	google/gemini-3.1-pro-preview	Gemini 3.1 Pro Preview	88.48%
4	Gemini 3.6 Flash	Google	88.08%
5	GPT-5.2 Codex	OpenAI	87.99%
6	Claude Opus 4.8	Anthropic	87.82%
7	Gemini 3.5 Flash	Google	87.60%
8	DeepSeek V4 Pro	DeepSeek	87.48%
9	Grok 4.5	xAI	87.35%
10	GPT-5.3 Codex	OpenAI	87.31%
11	Kimi K3	Moonshot AI	87.19%
12	Qwen3.7 Max	Alibaba	87.06%
13	Kimi K2.6	Moonshot AI	86.77%
14	GPT-5 Mini	OpenAI	86.61%
15	GPT-5.1	OpenAI	86.49%
16	Gemini 3 Pro Preview	Google	86.41%
17	Nemotron 3 Ultra 550b A55b	Nvidia	85.98%
18	Qwen3.6 Plus	Alibaba	85.95%
19	GPT-5.6 Terra	OpenAI	85.93%
20	GPT-5	OpenAI	85.91%
21	Muse Spark 1.1	Meta	85.91%
22	Gemini 3 Flash Preview	Google	85.59%
23	GPT-5.1 Codex	OpenAI	85.55%
24	Inkling	Thinkingmachines	85.52%
25	GPT-5.2	OpenAI	85.36%
26	Qwen3.5 Plus Thinking	Alibaba	85.33%
27	GPT-5.5	OpenAI	85.30%
28	Claude Opus 4.7	Anthropic	85.07%
29	GPT-5 Codex	OpenAI	84.72%
30	Claude Opus 4.6 Thinking	Anthropic	84.68%
31	Grok 4.3	xAI	84.49%
32	Grok 4.20 0309 Reasoning	xAI	84.27%
33	GPT-5.4	OpenAI	84.14%
34	GPT-5.4 Nano	OpenAI	84.01%
35	O3	OpenAI	83.91%
36	Kimi K2.5 Thinking	Moonshot AI	83.87%
37	Claude Opus 4.5 20251101 Thinking	Anthropic	83.67%
38	GPT-5.1 Codex Max	OpenAI	83.56%
39	Qwen3.5 Flash	Alibaba	83.28%
40	Grok 4 0709	xAI	83.25%
41	GPT Oss 120b	Fireworks AI	83.23%
42	GPT-5.6 Sol	OpenAI	82.60%
43	Claude Sonnet 5	Anthropic	82.43%
44	GLM 4.7	Zhipu AI	82.23%
45	O4 Mini	OpenAI	82.21%
46	MiniMax M3	MiniMax	82.15%
47	Claude Sonnet 4.6	Anthropic	82.09%
48	Kimi K2.7 Code	Moonshot AI	82.05%
49	GLM 5 Thinking	Zhipu AI	81.87%
50	MiniMax M2.1	MiniMax	81.76%
51	Mimo V2.5	Xiaomi	81.51%
52	GPT-5.4 Mini	OpenAI	81.47%
53	GLM 5.1	Zhipu AI	81.38%
54	Mimo V2.5 Pro	Xiaomi	81.35%
55	GLM 4.6	Zhipu AI	81.04%
56	DeepSeek V3p2 Thinking	Fireworks AI	80.69%
57	Grok 4.1 Fast Reasoning	xAI	80.64%
58	GPT Oss 20b	Fireworks AI	80.39%
59	Gemini 3.1 Flash Lite Preview	Google	80.12%
60	MiniMax M2.7	MiniMax	79.93%
61	MiniMax M2.5	MiniMax	79.21%
62	Gemini 2.5 Pro Preview 03 25	Google	79.16%
63	Gemini 3.5 Flash Lite	Google	79.01%
64	Grok 4 Fast Reasoning	xAI	78.97%
65	Qwen3 Max	Alibaba	78.22%
66	Grok 3 Mini Fast High Reasoning	xAI	76.22%
67	Gemini 2.5 Flash Preview 09 2025 Thinking	Google	76.21%
68	Gemini 2.5 Flash Preview 09 2025	Google	75.06%
69	Claude Opus 4.5	Anthropic	75.03%
70	Magistral Medium 2509	Mistral AI	74.86%
71	Claude Sonnet 4.5 20250929 Thinking	Anthropic	73.00%
72	Magistral Small 2509	Mistral AI	72.13%
73	O3 Mini	OpenAI	71.48%
74	Gemini 2.5 Flash Lite Preview 09 2025 Thinking	Google	71.39%
75	Qwen3 235b A22b	Fireworks AI	70.62%
76	Moonshotai Kimi K2 Instruct	Together AI	70.45%
77	DeepSeek R1	Fireworks AI	70.22%
78	GPT-5 Nano	OpenAI	70.22%
79	Claude Opus 4 20250514 Thinking	Anthropic	70.19%
80	DeepSeek V3p2	Fireworks AI	69.86%
81	GLM 5.2	Zhipu AI	69.50%
82	Gemini 2.5 Flash Lite Preview 09 2025	Google	67.67%
83	GLM 4.5	Zhipu AI	67.45%
84	Qwen3 Max Preview	Alibaba	66.91%
85	Laguna Xs.2	Poolside	66.61%
86	Claude Opus 4.1 20250805 Thinking	Anthropic	66.46%
87	Grok 3 Mini Fast Low Reasoning	xAI	66.27%
88	DeepSeek V3 0324	Fireworks AI	65.48%
89	Claude Opus 4.1	Anthropic	64.56%
90	Kimi K2 Thinking	Moonshot AI	63.15%
91	Claude Opus 4	Anthropic	62.63%
92	Claude Sonnet 4 20250514 Thinking	Anthropic	62.39%
93	Grok Code Fast 1	xAI	61.97%
94	Laguna M.1	Poolside	61.44%
95	Claude 3.7 Sonnet 20250219 Thinking	Anthropic	60.44%
96	Claude Sonnet 4	Anthropic	59.67%
97	Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561 Thinking	Together AI	58.37%
98	GPT-4.1 Mini	OpenAI	58.16%
99	Gemini 2.5 Flash Preview 04 17	Google	56.94%
100	Claude 3.7 Sonnet	Anthropic	56.66%
101	Mistral Large 2512	Mistral AI	55.34%
102	GPT-4.1	OpenAI	54.67%
103	Grok 3	xAI	52.90%
104	Devstral 2512	Mistral AI	51.84%
105	O1	OpenAI	50.26%
106	Claude 3.5 Sonnet	Anthropic	49.63%
107	Llama4 Maverick Instruct Basic	Fireworks AI	47.25%
108	Gemini 2.5 Flash Preview 04 17 Thinking	Google	46.87%
109	Grok 4 Fast Non Reasoning	xAI	46.09%
110	Mistral Medium 2505	Mistral AI	44.84%
111	Gemini 2.0 Flash 001	Google	43.61%
112	GPT-4o	OpenAI	43.44%
113	Labs Devstral Small 2512	Mistral AI	43.18%
114	GPT-4.1 Nano	OpenAI	42.72%
115	Grok 4.1 Fast Non Reasoning	xAI	42.62%
116	Claude 3.5 Haiku	Anthropic	41.92%
117	Gemini 1.5 Pro 002	Google	41.72%
118	Claude Haiku 4.5 20251001 Thinking	Anthropic	41.17%
119	Grok 2 1212	xAI	38.68%
120	Meta Llama Llama 4 Scout 17B 16E Instruct	Together AI	38.54%
121	Mistral Large 2411	Mistral AI	37.09%
122	Gemini 1.5 Flash 002	Google	36.91%
123	Meta Llama Llama 3.3 70B Instruct Turbo	Together AI	36.34%
124	Langston Nim Nvidia Llama 3.3 Nemotron Super 49b V1 42e84561	Together AI	36.31%
125	Command A 03 2025	Cohere	35.07%
126	Mistral Small 2503	Mistral AI	31.82%
127	GPT-4o Mini	OpenAI	26.42%
128	Jamba Large 1.6	AI21 Labs	22.32%
129	Command R Plus	Cohere	18.24%
130	Mistral Small 2402	Mistral AI	15.78%
131	Jamba Mini 1.6	AI21 Labs	9.92%