Coding

LiveCodeBench v6 leaderboard

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

As of 2026-10-07, the highest measured score on LiveCodeBench v6 is 93.2% by Sakana Fugu-Ultra. 157 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#	Model	Score	Source
1	Claude Opus 5.5	100.0%	estimated ± 2.1 pp, medium confidence
2	Gemini 4 Argon	100.0%	estimated ± 3.3 pp, medium confidence
3	MiMo-V2.6-Pro	100.0%	estimated ± 3.3 pp, medium confidence
4	Step 5 Preview	99.4%	estimated ± 3.3 pp, medium confidence
5	Claude Sonnet 5.5	99.0%	estimated ± 2.1 pp, medium confidence
6	Claude Fable 5.1	98.9%	estimated ± 2.1 pp, medium confidence
7	GPT-6 Sol	98.2%	estimated ± 3.3 pp, medium confidence
8	Grok 4.7	98.0%	estimated ± 3.3 pp, medium confidence
9	Claude Mythos 5	96.2%	estimated ± 2.1 pp, high confidence
10	Claude Fable 5	96.1%	estimated ± 2.1 pp, high confidence
11	Claude Haiku 5.5	95.9%	estimated ± 3.3 pp, medium confidence
12	Claude Opus 5	95.8%	estimated ± 2.1 pp, high confidence
13	GPT-6 Luna	95.5%	estimated ± 3.3 pp, medium confidence
14	GPT-6.1 Sol	95.2%	estimated ± 3.3 pp, medium confidence
15	Mistral Large 4	95.2%	estimated ± 3.3 pp, medium confidence
16	Ling 3.1 Flash	95.1%	estimated ± 3.3 pp, medium confidence
17	GPT-6 Astra	93.6%	estimated ± 3.1 pp, medium confidence
18	Grok 4.6	93.6%	estimated ± 3.1 pp, medium confidence
19	Gemini 3.8 Flash	93.5%	estimated ± 3.1 pp, medium confidence
20	Kimi K3	93.5%	estimated ± 3.1 pp, medium confidence
21	Gemini 3.7 Flash	93.4%	estimated ± 3.1 pp, medium confidence
22	Muse Spark 1.3	93.4%	estimated ± 3.1 pp, medium confidence
23	Sakana Fugu-Ultra	93.2%	measured
24	GLM-5.3	93.1%	estimated ± 3.1 pp, medium confidence
25	DeepSeek V4.1 Flash	93.1%	estimated ± 3.3 pp, medium confidence
26	Qwen3.8 Max	93.0%	estimated ± 2.1 pp, high confidence
27	Sakana Fugu	92.9%	measured
28	GLM-5.3-Flash	92.8%	estimated ± 3.3 pp, medium confidence
29	Qwen3.8-Omni-Flash	92.6%	measured
30	MiMo-V2.6-Flash	92.5%	estimated ± 3.3 pp, high confidence
31	Muse Spark 1.2	92.5%	estimated ± 3.1 pp, high confidence
32	Solar Open 2	92.4%	measured
33	Qwen3.8 Max Preview	92.3%	estimated ± 3.1 pp, high confidence
34	Claude Opus 4.8	92.2%	estimated ± 2.1 pp, high confidence
35	Hy4 preview	92.1%	estimated ± 2.1 pp, high confidence
36	Qwen3.8-Flash-Next	91.9%	measured
37	Gemini 3.6 Flash	91.7%	estimated ± 3.1 pp, high confidence
38	Grok 4.5	91.6%	estimated ± 2.1 pp, high confidence
39	GPT-5.6 Sol	91.6%	estimated ± 2.1 pp, high confidence
40	Gemini 3.1 Pro	91.5%	estimated ± 3.1 pp, high confidence
41	dots3-note Preview	91.5%	measured
42	GPT-5.6 Terra	91.1%	estimated ± 2.1 pp, high confidence
43	Beam	90.9%	estimated ± 2.1 pp, high confidence
44	GPT-5.6 Luna	90.8%	estimated ± 2.1 pp, high confidence
45	Ornith-1.5-397B	90.7%	estimated ± 2.1 pp, high confidence
46	Seed 2.1 Pro	90.6%	estimated ± 3.8 pp, low confidence
47	GLM-5.2	90.5%	estimated ± 2.1 pp, high confidence
48	Claude Opus 4.7 (Adaptive)	90.4%	estimated ± 2.1 pp, high confidence
49	Qwen3.8-27B	90.3%	measured
50	Muse Spark 1.1	90.2%	estimated ± 2.1 pp, high confidence
51	Ternary Bonsai 2 27B	90.1%	measured
52	Claude Sonnet 5	90.0%	estimated ± 2.1 pp, high confidence
53	Ornith-1.0-397B	89.6%	estimated ± 2.1 pp, high confidence
54	Kimi K2.6	89.6%	measured
55	Atria Dawn Preview	89.4%	estimated ± 2.1 pp, high confidence
56	Quasar 438B	89.3%	estimated ± 3.1 pp, high confidence
57	Laguna S 2.1	89.3%	estimated ± 2.1 pp, high confidence
58	Apodex 1.1	89.2%	estimated ± 3.1 pp, high confidence
59	Apodex 1.1 Mini	89.2%	estimated ± 3.1 pp, high confidence
60	Kimi K2.7 Code	89.2%	estimated ± 3.1 pp, high confidence
61	Qwen3.7 Max	89.0%	estimated ± 2.1 pp, high confidence
62	Nemotron 3 Ultra	89.0%	measured
63	GPT-5.5	89.0%	estimated ± 2.1 pp, high confidence
64	GLM-5.1	88.9%	estimated ± 2.1 pp, high confidence
65	Ornith-1.5-35B-A3B	88.6%	estimated ± 2.1 pp, high confidence
66	Hy3	88.6%	estimated ± 3.1 pp, high confidence
67	Hy3 Preview	88.6%	estimated ± 3.1 pp, high confidence
68	GPT-5.4	88.6%	estimated ± 2.1 pp, high confidence
69	Seed 2.1 Turbo	88.5%	estimated ± 3.8 pp, low confidence
70	MiniMax M3	88.4%	estimated ± 2.1 pp, high confidence
71	Qwen 3.6 Max (preview)	88.4%	estimated ± 2.1 pp, high confidence
72	MiMo-V2.5-Pro	88.3%	estimated ± 2.1 pp, high confidence
73	BTL-3	88.1%	measured
74	Ling 3.0 Flash	88.1%	estimated ± 2.1 pp, high confidence
75	Step 3.7 Flash	87.9%	estimated ± 2.1 pp, high confidence
76	MiniMax M2.7	87.9%	estimated ± 2.1 pp, high confidence
77	MiMo-V2.5	87.9%	estimated ± 2.1 pp, high confidence
78	Qwen3.7 Plus	87.8%	estimated ± 2.1 pp, high confidence
79	MAI-Thinking-1	87.7%	measured
80	GPT-5.4 mini	87.7%	estimated ± 3.1 pp, high confidence
81	GPT-5.4 nano	87.7%	estimated ± 3.1 pp, high confidence
82	GPT-5.3 Codex	87.5%	estimated ± 2.1 pp, high confidence
83	Gemini 3.5 Flash	87.4%	estimated ± 2.1 pp, high confidence
84	Inkling-Small	87.2%	estimated ± 2.1 pp, high confidence
85	Ember-1	87.1%	estimated ± 8.1 pp, low confidence
86	Qwen3.6 Plus	87.1%	measured
87	GPT-5.2	87.1%	estimated ± 2.1 pp, high confidence
88	Gemini 3.5 Flash-Lite	87.0%	estimated ± 2.1 pp, high confidence
89	DeepSeek V4 Pro 0813	87.0%	estimated ± 2.1 pp, high confidence
90	GLM-5	86.9%	estimated ± 2.1 pp, high confidence
91	Inkling	86.6%	estimated ± 2.1 pp, high confidence
92	Solar Pro 4	86.5%	estimated ± 3.3 pp, high confidence
93	Qwen3.6-27B	86.2%	estimated ± 2.1 pp, high confidence
94	Claude Opus 4.6	86.2%	estimated ± 2.1 pp, high confidence
95	Claude Sonnet 4.6	86.2%	estimated ± 8.1 pp, low confidence
96	Ling 3.0 Flash VL	86.1%	estimated ± 3.3 pp, high confidence
97	MiMo-V2-Pro	86.0%	estimated ± 8.1 pp, low confidence
98	Claude Sonnet 4.5	85.9%	estimated ± 8.1 pp, low confidence
99	DeepSeek V4 Flash 0731	85.9%	estimated ± 2.1 pp, high confidence
100	Muse Spark	85.8%	estimated ± 2.1 pp, high confidence
101	Ling 3.0 Flash FP8	85.7%	estimated ± 3.1 pp, high confidence
102	Grok 4.20	85.6%	estimated ± 2.1 pp, high confidence
103	MiMo-V2-Omni	85.5%	estimated ± 8.1 pp, low confidence
104	Claude 4.1 Opus	85.4%	estimated ± 8.1 pp, low confidence
105	MiMo-V2-Flash	85.4%	estimated ± 3.1 pp, high confidence
106	Muse Glimmer 30B	85.3%	estimated ± 2.1 pp, high confidence
107	GPT-5.1	85.2%	estimated ± 3.1 pp, high confidence
108	Claude Haiku 4.5	85.2%	estimated ± 8.1 pp, low confidence
109	Claude 4 Sonnet	85.1%	estimated ± 8.1 pp, low confidence
110	MAI-Code-1.1-Flash	85.1%	estimated ± 8.1 pp, low confidence
111	Qwen3.5-27B	85.0%	estimated ± 8.1 pp, low confidence
112	Ornith-1.0-35B	85.0%	estimated ± 2.1 pp, high confidence
113	Kimi K2.5	85.0%	measured
114	Claude Opus 4.5	84.8%	measured
115	Grok Code Fast 1	84.7%	estimated ± 8.1 pp, low confidence
116	Qwen3.6-35B-A3B	84.6%	estimated ± 2.1 pp, high confidence
117	Laguna M.1	84.5%	estimated ± 2.1 pp, high confidence
118	Mistral Medium 3.5 128B	84.3%	estimated ± 3.1 pp, high confidence
119	Qwen3.5-35B-A3B	84.3%	estimated ± 8.1 pp, low confidence
120	Kimi K2.5 (Reasoning)	84.2%	estimated ± 3.1 pp, high confidence
121	K-EXAONE 2.0	84.1%	estimated ± 3.3 pp, high confidence
122	A.X K2	84.0%	measured
123	Laguna XS 2.1	83.9%	estimated ± 2.1 pp, high confidence
124	Ornith-1.5-9B	83.8%	estimated ± 2.1 pp, high confidence
125	Qwen3.5-122B-A10B	83.8%	estimated ± 3.1 pp, high confidence
126	GLM-4.7	83.6%	estimated ± 3.1 pp, high confidence
127	Qwen3.5 397B	83.6%	measured
128	Laguna XS.2	83.3%	estimated ± 2.1 pp, high confidence
129	Gemma 4 31B	82.8%	estimated ± 3.1 pp, high confidence
130	Grok 4.3	82.3%	estimated ± 3.1 pp, high confidence
131	Ornith-1.0-9B	81.9%	estimated ± 2.1 pp, high confidence
132	DeepSeek V3 0324	81.4%	estimated ± 3.3 pp, high confidence
133	North Mini Code	81.3%	estimated ± 3.3 pp, high confidence
134	o1	81.2%	estimated ± 3.1 pp, high confidence
135	Gemma 4 26B A4B	81.0%	estimated ± 3.1 pp, high confidence
136	LongCat-Flash-Lite-Sparse	81.0%	estimated ± 2.1 pp, high confidence
137	Mercury 2.5	81.0%	estimated ± 3.3 pp, high confidence
138	GPT-5 (high)	80.3%	estimated ± 3.1 pp, high confidence
139	Nemotron 3 Super 100B	80.2%	estimated ± 3.1 pp, high confidence
140	o1-preview	78.4%	estimated ± 3.1 pp, high confidence
141	Gemini 2.5 Pro	78.0%	estimated ± 3.1 pp, high confidence
142	K-Exaone	77.4%	estimated ± 3.1 pp, high confidence
143	GPT-4.1	76.7%	estimated ± 8.1 pp, low confidence
144	LLaDA2.2-flash	76.5%	estimated ± 2.1 pp, high confidence
145	GPT-OSS 120B	76.5%	estimated ± 3.1 pp, high confidence
146	Granite 4.2 30B	75.8%	measured
147	Command A+	75.0%	estimated ± 3.1 pp, high confidence
148	Nemotron 3.5 Lightning 30B A3B NVFP4	74.3%	estimated ± 3.1 pp, high confidence
149	Mistral Small 4	74.2%	estimated ± 3.1 pp, high confidence
150	Mistral Small 4 (Reasoning)	74.2%	estimated ± 3.1 pp, high confidence
151	Trinity-Large-Preview	73.7%	estimated ± 3.1 pp, high confidence
152	Trinity-Large-Thinking	73.7%	estimated ± 3.1 pp, high confidence
153	Ling 2.6 Flash	73.4%	estimated ± 3.1 pp, high confidence
154	Granite 4.2 8B	73.2%	measured
155	Gemini 1.5 Pro	72.3%	estimated ± 3.1 pp, high confidence
156	Gemma 4 12B	72.0%	measured
157	DeepSeek V3	71.9%	estimated ± 3.1 pp, high confidence
158	o3-mini	71.7%	estimated ± 8.1 pp, low confidence
159	Claude 3.5 Sonnet	71.4%	estimated ± 8.1 pp, low confidence
160	GPT-4 Turbo	70.8%	estimated ± 3.1 pp, high confidence
161	GPT-OSS 20B	70.3%	estimated ± 3.1 pp, high confidence
162	GPT-4.1 mini	69.9%	estimated ± 3.1 pp, high confidence
163	Mellum2-12B-A2.5B-Thinking	69.9%	measured
164	Mistral Large 3	69.8%	estimated ± 3.1 pp, high confidence
165	Granite 4.2 3B	69.7%	measured
166	Claude 3 Opus	69.5%	estimated ± 3.1 pp, high confidence
167	Solar Pro 3	69.2%	estimated ± 3.3 pp, high confidence
168	MiniCPM5-2B	69.1%	measured
169	Ling 3.0 Tiny	68.1%	estimated ± 3.3 pp, high confidence
170	Llama 4 Maverick	67.0%	estimated ± 3.1 pp, high confidence
171	BTL-4	66.1%	measured
172	ZAYA1-8B	65.8%	measured
173	ZAYA1-74B-Preview	65.7%	measured
174	Celeris-1	65.4%	estimated ± 3.1 pp, high confidence
175	Nemotron 3 Nano 30B	65.4%	estimated ± 3.1 pp, high confidence
176	Nemotron 3 Nano Omni 30B A3B	64.9%	estimated ± 3.1 pp, high confidence
177	Ultravox v0.6 Llama 3.3 70B	63.3%	estimated ± 3.1 pp, high confidence
178	GPT-4o mini	62.8%	estimated ± 3.1 pp, high confidence
179	GPT-4.1 nano	62.6%	estimated ± 3.1 pp, high confidence
180	Gemma 3 27B	61.6%	estimated ± 3.1 pp, high confidence
181	Gemma 4 E4B	60.9%	estimated ± 3.1 pp, high confidence
182	Llama 4 Scout	59.7%	estimated ± 3.1 pp, high confidence
183	Agents-A1-4B	59.6%	measured
184	LFM2.5-2.6B	59.4%	measured
185	Gemma 4 E2B	58.8%	estimated ± 3.1 pp, medium confidence
186	Mellum2-12B-A2.5B-Instruct	37.2%	measured
187	MiniCPM5-1B	33.5%	measured
188	LLaDA2.2-mini	28.1%	measured

LiveCodeBench v6 leaderboard

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General