benchgap
Coding

LiveCodeBench v6 leaderboard

As of 2026-11-21, the highest measured score on LiveCodeBench v6 is 93.2% by Sakana Fugu-Ultra. 164 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#ModelScoreSource
1Claude Fable 5.2100.0%estimated ± 2.7 pp, medium confidence
2Claude Opus 5.5100.0%estimated ± 2.7 pp, medium confidence
3Gemini 4 Argon100.0%estimated ± 4.9 pp, medium confidence
4Kimi K3.1 Pro100.0%estimated ± 4.9 pp, medium confidence
5MiMo-V2.6-Pro100.0%estimated ± 4.9 pp, medium confidence
6Muse Wonder100.0%estimated ± 4.9 pp, medium confidence
7Muse Wonder Full100.0%estimated ± 4.9 pp, medium confidence
8Muse Spark Open98.9%estimated ± 4.9 pp, medium confidence
9GPT-6 Sol98.1%estimated ± 4.9 pp, medium confidence
10Grok 4.797.9%estimated ± 4.9 pp, medium confidence
11Claude Sonnet 5.597.3%estimated ± 2.7 pp, medium confidence
12Claude Mythos 596.9%estimated ± 2.7 pp, medium confidence
13DeepSeek V4.1 Flash 101796.9%estimated ± 1.9 pp, medium confidence
14ERNIE 5.5 Flash Preview96.1%estimated ± 1.9 pp, medium confidence
15Qwen3.8 Max95.8%estimated ± 1.7 pp, medium confidence
16GPT-6.1 Terra95.6%estimated ± 4.9 pp, medium confidence
17Claude Haiku 5.595.5%estimated ± 4.9 pp, medium confidence
18MiniMax M3.1 Flash95.3%estimated ± 1.9 pp, medium confidence
19LongCat-2.595.1%estimated ± 1.9 pp, medium confidence
20GPT-6 Luna95.1%estimated ± 4.9 pp, medium confidence
21GPT-6.1 Sol94.7%estimated ± 4.9 pp, medium confidence
22Hy4 preview94.6%estimated ± 1.7 pp, medium confidence
23GLM-5.394.4%estimated ± 1.9 pp, medium confidence
24Ornith-1.5-397B94.2%estimated ± 1.7 pp, medium confidence
25GLM-5.3-Flash94.0%estimated ± 1.9 pp, medium confidence
26Sakana Fugu-Ultra93.2%measured
27Step 5 Preview93.1%estimated ± 4.4 pp, high confidence
28Sakana Fugu92.9%measured
29Inkling 2 Small92.8%estimated ± 4.9 pp, medium confidence
30Ling 3.1 Flash92.8%estimated ± 2.7 pp, high confidence
31Qwen3.8-Omni-Flash92.6%measured
32Solar Open 292.4%measured
33Qwen3.8-Flash-Next91.9%measured
34MiMo-V2.6-Flash91.8%estimated ± 4.9 pp, high confidence
35Beam 1.0 501B-A23B91.8%estimated ± 2.7 pp, high confidence
36Beam 1.0 501B-A23B FP891.8%estimated ± 2.7 pp, high confidence
37Beam 1.0 501B-A23B NVFP491.7%estimated ± 2.7 pp, high confidence
38Qwen3.7 Max91.6%measured
39Ornith-1.0-397B91.6%estimated ± 1.7 pp, high confidence
40GLM-5.291.6%estimated ± 1.7 pp, high confidence
41dots3-note Preview91.5%measured
42Claude Fable 5.191.1%estimated ± 2.2 pp, low confidence
43Seed 2.1 Pro90.9%estimated ± 1.9 pp, high confidence
44Claude Fable 590.8%estimated ± 2.2 pp, low confidence
45Gemini 3.8 Flash90.7%estimated ± 2.2 pp, low confidence
46GPT-5.6 Luna90.6%estimated ± 2.7 pp, high confidence
47Claude Opus 590.5%estimated ± 2.2 pp, low confidence
48Gemini 3.7 Flash90.3%estimated ± 2.2 pp, low confidence
49Qwen3.8-27B90.3%measured
50Gemini 3.1 Pro90.3%estimated ± 2.2 pp, low confidence
51Grok 4.690.1%estimated ± 2.2 pp, low confidence
52Ripple-65B Preview90.1%estimated ± 2.7 pp, high confidence
53Gemini 3.6 Flash90.1%estimated ± 2.2 pp, low confidence
54Ternary Bonsai 2 27B90.1%measured
55GPT-5.2-Codex90.0%estimated ± 2.2 pp, low confidence
56Sage-A1.590.0%estimated ± 2.7 pp, high confidence
57Claude Opus 4.889.9%estimated ± 2.2 pp, low confidence
58Gemini 3.5 Flash89.9%estimated ± 2.2 pp, low confidence
59Grok 4.589.8%estimated ± 2.2 pp, low confidence
60GPT-5.3 Codex89.7%estimated ± 2.2 pp, low confidence
61Ornith-1.5-35B-A3B89.7%estimated ± 1.7 pp, high confidence
62Kimi K389.7%estimated ± 2.2 pp, low confidence
63Kimi K2.689.6%measured
64Qwen3.7 Plus89.6%measured
65Seed 2.1 Turbo89.5%estimated ± 1.9 pp, high confidence
66Atria Dawn Preview89.4%estimated ± 2.7 pp, high confidence
67GPT-5.189.4%estimated ± 2.2 pp, medium confidence
68Gemini 3 Pro89.3%estimated ± 2.2 pp, medium confidence
69Laguna S 2.189.3%estimated ± 2.7 pp, high confidence
70GPT-589.1%estimated ± 2.2 pp, medium confidence
71GPT-5.6 Terra89.1%estimated ± 2.2 pp, medium confidence
72Inkling-Small89.1%estimated ± 2.2 pp, medium confidence
73Muse Spark 1.189.1%estimated ± 2.2 pp, medium confidence
74MiniMax M389.1%estimated ± 1.7 pp, high confidence
75Nemotron 3 Ultra89.0%measured
76Gemini 3 Flash89.0%estimated ± 2.2 pp, medium confidence
77GPT-5.1-Codex89.0%estimated ± 2.2 pp, medium confidence
78Inkling88.9%estimated ± 2.2 pp, medium confidence
79GPT-6 Astra88.9%estimated ± 4.7 pp, medium confidence
80GPT-5.288.9%estimated ± 2.2 pp, medium confidence
81GPT-5.588.8%estimated ± 2.2 pp, medium confidence
82Muse Spark 1.388.8%estimated ± 4.7 pp, medium confidence
83Claude Opus 4.788.8%estimated ± 2.2 pp, medium confidence
84GLM-5.188.7%estimated ± 1.7 pp, high confidence
85Solar Pro 488.6%estimated ± 4.9 pp, high confidence
86Claude Opus 4.6 (Adaptive)88.6%estimated ± 2.2 pp, medium confidence
87Grok 4.388.5%estimated ± 2.2 pp, medium confidence
88Grok 4.2088.4%estimated ± 2.2 pp, medium confidence
89Muse Spark 1.288.4%estimated ± 4.7 pp, high confidence
90Qwen3.8 Max Preview88.3%estimated ± 4.7 pp, high confidence
91GPT-5.488.3%estimated ± 2.2 pp, medium confidence
92GPT-5.4 nano88.3%estimated ± 2.2 pp, medium confidence
93Ling 3.0 Flash88.3%estimated ± 2.2 pp, medium confidence
94Nemotron 3.5 Super VL 120B A12B88.2%measured
95o388.2%estimated ± 2.2 pp, medium confidence
96Step 3.7 Flash88.1%estimated ± 2.7 pp, high confidence
97Claude Opus 4.5 Thinking88.1%estimated ± 2.2 pp, medium confidence
98BTL-388.1%measured
99GPT-5.1-Codex-Max88.1%estimated ± 2.2 pp, medium confidence
100Qwen 3.6 Max (preview)88.0%estimated ± 1.7 pp, high confidence
101Qwen3.5 Flash88.0%estimated ± 2.2 pp, medium confidence
102Grok 487.9%estimated ± 2.2 pp, medium confidence
103DeepSeek V4 Pro87.8%estimated ± 1.7 pp, medium confidence
104MAI-Thinking-187.7%measured
105GPT-5.6 Sol87.6%estimated ± 2.2 pp, medium confidence
106Claude Sonnet 587.5%estimated ± 2.2 pp, medium confidence
107o4-mini87.5%estimated ± 2.2 pp, medium confidence
108Nemotron 3.5 Super VL 120B A12B NVFP487.4%measured
109Claude Sonnet 4.687.4%estimated ± 2.2 pp, medium confidence
110Kimi K2.7 Code87.4%estimated ± 2.2 pp, medium confidence
111Spark-X2.6-4B87.4%estimated ± 2.7 pp, high confidence
112GPT-5.4 mini87.1%estimated ± 2.2 pp, medium confidence
113MiMo-V2.587.1%estimated ± 2.2 pp, medium confidence
114Qwen3.6 Plus87.1%measured
115MiMo-V2.5-Pro87.1%estimated ± 2.2 pp, medium confidence
116MiniMax M2.787.0%estimated ± 1.7 pp, high confidence
117Claude Opus 4.687.0%estimated ± 2.7 pp, high confidence
118Quasar 438B87.0%estimated ± 4.7 pp, high confidence
119GLM-4.686.9%estimated ± 2.2 pp, medium confidence
120Apodex 1.186.9%estimated ± 4.7 pp, high confidence
121Apodex 1.1 Mini86.9%estimated ± 4.7 pp, high confidence
122Grok 4.1 Fast (Reasoning)86.7%estimated ± 2.2 pp, medium confidence
123Muse Spark86.6%estimated ± 2.7 pp, high confidence
124Hy386.6%estimated ± 4.7 pp, high confidence
125Gemini 3.1 Flash-Lite86.5%estimated ± 2.2 pp, medium confidence
126Step 3.5 Flash86.4%measured
127Muse Glimmer 30B86.2%estimated ± 2.7 pp, high confidence
128GLM-586.1%estimated ± 1.7 pp, high confidence
129Gemini 3.5 Flash-Lite86.0%estimated ± 2.2 pp, medium confidence
130Grok 4 Fast (Reasoning)86.0%estimated ± 2.2 pp, medium confidence
131Qwen3 Max85.6%estimated ± 2.2 pp, medium confidence
132DeepSeek V4 Flash 073185.5%estimated ± 1.7 pp, medium confidence
133Kimi K2.585.0%measured
134GLM-4.784.9%measured
135Ling 3.0 Flash VL84.8%estimated ± 4.9 pp, high confidence
136Claude Opus 4.584.8%measured
137Laguna XS 2.184.7%estimated ± 2.7 pp, high confidence
138Mistral Medium 3.5 128B84.2%estimated ± 4.7 pp, high confidence
139A.X K284.0%measured
140Qwen3.6-27B83.9%measured
141Qwen3.5 397B83.6%measured
142Claude Sonnet 4.5 Thinking83.2%estimated ± 2.2 pp, low confidence
143Ornith-1.0-35B82.8%estimated ± 1.7 pp, high confidence
144o3-mini82.5%estimated ± 2.2 pp, low confidence
145o182.2%estimated ± 4.7 pp, high confidence
146Mellum2.1-12B-A2.5B-Thinking82.0%measured
147LongCat-Flash-Lite-Sparse81.9%estimated ± 2.7 pp, high confidence
148Laguna M.180.8%estimated ± 2.2 pp, low confidence
149Hy3 Preview80.8%estimated ± 4.4 pp, high confidence
150Ornith-1.5-9B80.7%estimated ± 1.7 pp, medium confidence
151Qwen3.5-27B80.7%measured
152Laguna XS.280.6%estimated ± 2.2 pp, low confidence
153MiMo-V2-Flash80.6%measured
154GLM-4.580.4%estimated ± 2.2 pp, low confidence
155Qwen3.6-35B-A3B80.4%measured
156Nemotron 3.5 Bolt 74B-A9B80.4%estimated ± 4.9 pp, high confidence
157Flint80.3%measured
158o1-preview80.1%estimated ± 4.7 pp, high confidence
159Gemma 4 31B80.0%measured
160Ling 3.0 Flash FP879.8%estimated ± 4.4 pp, high confidence
161Gemini 2.5 Pro79.8%estimated ± 4.7 pp, high confidence
162DeepSeek V3 032479.7%estimated ± 4.9 pp, high confidence
163North Mini Code79.5%estimated ± 4.9 pp, high confidence
164K-Exaone79.3%estimated ± 4.7 pp, high confidence
165Qwen3.5-122B-A10B78.9%measured
166GPT-OSS 120B78.5%estimated ± 4.7 pp, high confidence
167LLaDA2.2-flash77.4%estimated ± 2.7 pp, high confidence
168Ornith-1.0-9B77.3%estimated ± 1.7 pp, medium confidence
169Mercury 2.577.1%estimated ± 4.4 pp, high confidence
170Gemma 4 26B A4B77.1%measured
171Command A+77.1%estimated ± 4.7 pp, high confidence
172K-EXAONE 2.076.5%estimated ± 4.4 pp, high confidence
173Mistral Small 476.4%estimated ± 4.7 pp, high confidence
174Trinity-Large-Preview75.8%estimated ± 4.7 pp, high confidence
175Trinity-Large-Thinking75.8%estimated ± 4.7 pp, high confidence
176Granite 4.2 30B75.8%measured
177Qwen3.5-35B-A3B74.6%measured
178Gemini 1.5 Pro74.3%estimated ± 4.7 pp, high confidence
179DeepSeek V373.9%estimated ± 4.7 pp, high confidence
180LFM3-1.7B73.5%measured
181Granite 4.2 8B73.2%measured
182GPT-4 Turbo72.6%estimated ± 4.7 pp, high confidence
183Gemma 4 12B72.0%measured
184GPT-OSS 20B71.9%estimated ± 4.7 pp, high confidence
185LFM2.5-24B-A2B71.7%estimated ± 4.4 pp, high confidence
186GPT-4.1 mini71.5%estimated ± 4.7 pp, high confidence
187Mistral Large 371.3%estimated ± 4.7 pp, high confidence
188Nemotron 3 Nano Omni 30B A3B71.1%estimated ± 4.4 pp, high confidence
189Claude 3 Opus70.8%estimated ± 4.7 pp, high confidence
190Nemotron 3.5 Lightning 30B A3B NVFP470.6%estimated ± 4.4 pp, high confidence
191Mellum2-12B-A2.5B-Thinking69.9%measured
192Granite 4.2 3B69.7%measured
193MiniCPM5-2B69.1%measured
194Mistral Small 3.168.6%estimated ± 4.9 pp, high confidence
195Nemotron 3 Nano 30B68.3%measured
196Ling 2.6 Flash67.9%estimated ± 4.4 pp, high confidence
197Llama 4 Maverick67.2%estimated ± 4.7 pp, high confidence
198Solar Pro 366.4%estimated ± 4.9 pp, high confidence
199BTL-466.1%measured
200ZAYA1-8B65.8%measured
201ZAYA1-74B-Preview65.7%measured
202Ling 3.0 Tiny65.1%estimated ± 4.9 pp, high confidence
203Celeris-164.6%estimated ± 4.7 pp, high confidence
204GLM-4.7-Flash64.0%measured
205Ultravox v0.6 Llama 3.3 70B60.3%estimated ± 4.7 pp, high confidence
206Agents-A1-4B59.6%measured
207LFM2.5-2.6B59.4%measured
208GPT-4o mini59.3%estimated ± 4.7 pp, high confidence
209GPT-4.1 nano58.8%estimated ± 4.7 pp, high confidence
210Gemma 3 27B56.4%estimated ± 4.7 pp, high confidence
211Kimi K253.7%measured
212Gemma 4 E4B52.0%measured
213Qwen3 235B 250751.8%measured
214Llama 4 Scout51.4%estimated ± 4.7 pp, high confidence
215Magistral Medium50.3%measured
216Magistral Small47.4%measured
217Gemma 4 E2B44.0%measured
218Mellum2-12B-A2.5B-Instruct37.2%measured
219MiniCPM5-1B33.5%measured
220Mistral Medium 330.3%measured
221LLaDA2.2-mini28.1%measured

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General