benchgap
Coding

FrontierCode 1.1 Main leaderboard

As of 2026-11-21, the highest measured score on FrontierCode 1.1 Main is 54.4% by Claude Opus 5.5. 191 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#ModelScoreSource
1Claude Fable 5.157.9%estimated ± 1.6 pp, low confidence
2GPT-6.2 Astra55.0%estimated ± 3.0 pp, medium confidence
3Claude Mythos 554.9%estimated ± 4.2 pp, medium confidence
4Grok 4.654.7%estimated ± 1.6 pp, low confidence
5Muse Wonder Full54.5%estimated ± 3.5 pp, low confidence
6Claude Opus 5.554.4%measured
7Gemini 4 Argon53.8%estimated ± 2.8 pp, medium confidence
8MiniMax M3.1 Flash53.8%estimated ± 2.8 pp, medium confidence
9Gemini 4 Carbon53.6%estimated ± 3.0 pp, medium confidence
10Claude Fable 553.5%measured
11DeepSeek V4.1 Flash 101753.5%estimated ± 3.5 pp, low confidence
12Claude Opus 553.4%measured
13GPT-6 Astra53.3%measured
14Gemini 3.8 Flash52.8%estimated ± 1.6 pp, medium confidence
15Muse Spark 1.352.6%estimated ± 3.5 pp, low confidence
16Ember-152.5%estimated ± 3.5 pp, low confidence
17GPT-6.1 Sol52.4%estimated ± 3.0 pp, medium confidence
18Claude Fable 5.252.1%estimated ± 3.5 pp, low confidence
19Gemini 4 Neon51.9%estimated ± 3.8 pp, low confidence
20Pareto 26.951.9%estimated ± 3.5 pp, low confidence
21GPT-6.2 Sol51.7%estimated ± 3.9 pp, medium confidence
22MAI Code 251.5%estimated ± 3.5 pp, low confidence
23Particle-151.2%estimated ± 3.5 pp, low confidence
24GLM-5.3-Flash51.1%estimated ± 3.0 pp, medium confidence
25Gemini 4 Hydrogen51.1%estimated ± 3.8 pp, low confidence
26Qwen4 Max51.1%estimated ± 3.0 pp, medium confidence
27Kimi K3.1 Pro50.9%estimated ± 3.0 pp, medium confidence
28MiMo-V2.6-Pro50.8%estimated ± 3.5 pp, low confidence
29GPT-5.6 Sol50.5%estimated ± 1.6 pp, medium confidence
30ERNIE 5.5 Flash Preview50.4%estimated ± 3.5 pp, low confidence
31Grok 4.750.3%estimated ± 3.5 pp, low confidence
32Hy450.1%estimated ± 3.0 pp, medium confidence
33SWE-250.0%measured
34Grok 4.549.9%estimated ± 1.6 pp, medium confidence
35Pareto 26.10 Preview49.8%estimated ± 3.5 pp, low confidence
36GLM-5.349.6%estimated ± 3.0 pp, medium confidence
37GPT-6 Sol49.2%estimated ± 3.5 pp, low confidence
38MiMo-V2.6-Flash48.7%estimated ± 3.5 pp, low confidence
39Step 5 Preview48.6%estimated ± 3.5 pp, low confidence
40GPT-6 Luna48.0%estimated ± 3.5 pp, low confidence
41GPT-5.6 Terra47.9%estimated ± 1.6 pp, medium confidence
42Inkling 2 Small47.8%estimated ± 3.0 pp, medium confidence
43Hy4 flash46.9%estimated ± 3.5 pp, low confidence
44LongCat-2.546.9%estimated ± 3.5 pp, low confidence
45Hy4 preview46.7%estimated ± 3.5 pp, low confidence
46Sakana Fugu-Ultra46.7%estimated ± 5.0 pp, high confidence
47GPT-6.1 Terra46.6%estimated ± 6.6 pp, medium confidence
48Claude Opus 4.846.5%measured
49Claude Haiku 5.546.4%measured
50Claude Sonnet 5.546.2%measured
51DeepSeek V4 Pro45.8%estimated ± 3.5 pp, low confidence
52Mistral Large 445.7%estimated ± 3.9 pp, medium confidence
53SWE-2 Flash45.5%measured
54Ultramarine 1 Flash45.1%estimated ± 3.5 pp, low confidence
55Muse Spark Open44.3%estimated ± 3.5 pp, low confidence
56MAI Code 2 Flash44.2%estimated ± 3.5 pp, low confidence
57Ling 3.1 Flash44.1%estimated ± 3.5 pp, low confidence
58GPT-5.6 Luna43.9%estimated ± 1.6 pp, medium confidence
59Muse Spark 1.243.9%estimated ± 3.5 pp, low confidence
60Gemini 3.7 Flash43.6%measured
61Kimi K343.6%estimated ± 1.6 pp, medium confidence
62Qwen3.8-Flash-Next43.5%estimated ± 3.5 pp, low confidence
63GPT-5.543.0%measured
64Qwen3.8-Omni-Flash43.0%estimated ± 3.5 pp, low confidence
65Claude Sonnet 542.7%measured
66SWE-1.742.3%measured
67Qwen3.8 Max42.3%estimated ± 3.5 pp, low confidence
68Ornith-1.5-397B41.9%estimated ± 3.5 pp, low confidence
69Qwen3.8 Max Preview41.6%estimated ± 4.6 pp, high confidence
70Muse Wonder41.3%estimated ± 3.0 pp, medium confidence
71GPT-5.1-Codex41.2%estimated ± 6.2 pp, medium confidence
72DeepSeek V4 Flash 073141.0%estimated ± 3.5 pp, low confidence
73Gemini 3.1 Pro40.7%estimated ± 2.8 pp, low confidence
74GPT-5.440.7%estimated ± 2.8 pp, low confidence
75Sage-A1.540.4%estimated ± 5.0 pp, high confidence
76Muse Spark 1.140.3%estimated ± 3.5 pp, low confidence
77Atria Dawn Preview39.5%estimated ± 5.0 pp, high confidence
78Laguna S 2.139.4%estimated ± 5.0 pp, high confidence
79Sakana Fugu39.2%estimated ± 5.0 pp, high confidence
80Composer 2.538.9%estimated ± 1.6 pp, low confidence
81Claude Opus 4.738.5%measured
82Qwen 3.6 Max (preview)38.3%estimated ± 5.0 pp, high confidence
83o338.0%estimated ± 6.2 pp, medium confidence
84GLM-5.237.9%estimated ± 1.6 pp, low confidence
85MiMo-V2.537.6%estimated ± 5.0 pp, high confidence
86GPT-5.1-Codex-Max37.4%estimated ± 6.2 pp, medium confidence
87Inkling 2 Mini37.2%estimated ± 3.5 pp, low confidence
88Spark-X2.6-4B36.6%estimated ± 5.0 pp, high confidence
89Qwen3.8-27B36.6%estimated ± 4.6 pp, high confidence
90Gemini 3.6 Flash36.5%estimated ± 1.6 pp, low confidence
91GPT-5.3 Codex36.3%estimated ± 4.2 pp, medium confidence
92Mercury 336.0%estimated ± 3.5 pp, low confidence
93o4-mini34.9%estimated ± 6.2 pp, medium confidence
94Kimi K2.7 Code33.0%estimated ± 1.6 pp, low confidence
95GLM-4.632.7%estimated ± 6.2 pp, low confidence
96Claude Opus 4.5 Thinking32.5%estimated ± 5.2 pp, medium confidence
97Gemini 3 Pro32.5%estimated ± 5.2 pp, medium confidence
98Gemini 3.5 Flash32.2%estimated ± 1.6 pp, low confidence
99Grok 4.1 Fast (Reasoning)32.0%estimated ± 6.2 pp, low confidence
100Ornith-1.0-397B31.7%estimated ± 4.2 pp, medium confidence
101Gemini 3 Flash30.6%estimated ± 5.2 pp, medium confidence
102Beam 1.0 501B-A23B30.3%estimated ± 4.2 pp, medium confidence
103Grok 4 Fast (Reasoning)29.3%estimated ± 6.2 pp, low confidence
104Claude Opus 4.529.1%estimated ± 4.2 pp, medium confidence
105Quasar 438B29.1%estimated ± 4.6 pp, high confidence
106Beam 1.0 501B-A23B NVFP428.7%estimated ± 4.2 pp, medium confidence
107Apodex 1.1 Mini28.7%estimated ± 4.6 pp, high confidence
108MiniMax M328.4%estimated ± 4.2 pp, medium confidence
109Qwen3.7 Max28.2%estimated ± 4.2 pp, medium confidence
110MiMo-V2.5-Pro28.1%estimated ± 4.6 pp, high confidence
111Qwen3 Max28.0%estimated ± 6.2 pp, low confidence
112Inkling-Small27.9%estimated ± 4.2 pp, medium confidence
113Kimi K2.627.9%estimated ± 4.2 pp, medium confidence
114Beam 1.0 501B-A23B FP827.9%estimated ± 4.2 pp, medium confidence
115GPT-5.227.5%estimated ± 4.2 pp, medium confidence
116Claude Opus 4.6 (Adaptive)27.0%estimated ± 4.3 pp, medium confidence
117GPT-5.4 mini27.0%measured
118Claude Opus 4.626.9%measured
119Hy326.9%estimated ± 4.6 pp, high confidence
120Ornith-1.5-35B-A3B25.7%estimated ± 4.2 pp, low confidence
121Qwen3.6 Plus25.4%estimated ± 4.2 pp, low confidence
122BTL-424.7%estimated ± 4.2 pp, low confidence
123dots3-note Preview24.7%estimated ± 4.2 pp, low confidence
124Claude Sonnet 4.624.3%measured
125Granite 4.2 30B24.2%estimated ± 5.0 pp, medium confidence
126MiMo-V2-Pro24.0%estimated ± 4.2 pp, low confidence
127Claude Sonnet 4.5 Thinking23.8%estimated ± 5.2 pp, low confidence
128GLM-523.6%estimated ± 4.2 pp, low confidence
129Apodex 1.123.4%estimated ± 4.2 pp, low confidence
130Qwen3.7 Plus23.4%estimated ± 4.2 pp, low confidence
131Mistral Medium 3.5 128B23.3%estimated ± 4.2 pp, low confidence
132Muse Spark22.9%estimated ± 4.2 pp, low confidence
133Claude Sonnet 4.522.6%estimated ± 4.2 pp, low confidence
134Qwen3.6-27B22.6%estimated ± 4.2 pp, low confidence
135Inkling22.4%estimated ± 3.8 pp, low confidence
136LLaDA2.2-flash22.1%estimated ± 5.0 pp, medium confidence
137Kimi K2.521.8%estimated ± 4.2 pp, low confidence
138Grok 4.2021.7%estimated ± 4.2 pp, low confidence
139Ripple-65B Preview21.3%estimated ± 4.2 pp, low confidence
140Qwen3.5 397B20.8%estimated ± 4.2 pp, low confidence
141Mellum2.1-12B-A2.5B-Thinking20.7%estimated ± 5.0 pp, medium confidence
142Muse Glimmer 30B20.4%estimated ± 4.2 pp, low confidence
143Ling 3.0 Flash20.4%estimated ± 4.6 pp, medium confidence
144Ornith-1.0-35B19.7%estimated ± 4.2 pp, low confidence
145GPT-5.119.6%estimated ± 4.6 pp, medium confidence
146Gemini 3.5 Flash-Lite19.5%estimated ± 4.6 pp, medium confidence
147MiMo-V2-Omni18.3%estimated ± 4.2 pp, low confidence
148Laguna M.118.0%estimated ± 4.2 pp, low confidence
149Claude 4.1 Opus17.8%estimated ± 4.2 pp, low confidence
150Hy3 Preview17.6%estimated ± 4.2 pp, low confidence
151Step 3.5 Flash17.6%estimated ± 4.2 pp, low confidence
152GPT-5.2-Codex17.3%estimated ± 4.3 pp, low confidence
153Flint17.3%estimated ± 4.2 pp, low confidence
154Qwen3.5 Flash16.6%estimated ± 5.2 pp, low confidence
155GLM-4.716.5%estimated ± 4.2 pp, low confidence
156MAI-Thinking-116.0%estimated ± 4.2 pp, low confidence
157Gemma 4 31B15.9%estimated ± 4.6 pp, medium confidence
158MiMo-V2-Flash15.8%estimated ± 4.2 pp, low confidence
159Qwen3.6-35B-A3B15.8%estimated ± 4.2 pp, low confidence
160Claude Haiku 4.515.7%estimated ± 4.2 pp, low confidence
161Grok 4.315.2%estimated ± 4.6 pp, medium confidence
162Gemini 3.1 Flash-Lite14.8%estimated ± 5.2 pp, low confidence
163Jais-3-MoVA-36B-A4B14.8%estimated ± 4.2 pp, low confidence
164Granite 4.2 8B14.6%estimated ± 5.0 pp, medium confidence
165Claude 4 Sonnet14.6%estimated ± 4.2 pp, low confidence
166MAI-Code-1.1-Flash14.4%estimated ± 4.2 pp, low confidence
167Claude 4 Opus14.3%estimated ± 4.2 pp, low confidence
168Jais-3-32B14.1%estimated ± 4.2 pp, low confidence
169Qwen3.5-27B14.1%estimated ± 4.2 pp, low confidence
170o113.8%estimated ± 4.6 pp, medium confidence
171GLM-5.113.8%estimated ± 4.3 pp, low confidence
172Step 3.7 Flash13.8%estimated ± 4.6 pp, medium confidence
173Gemma 4 26B A4B13.6%estimated ± 4.6 pp, medium confidence
174Qwen3.5-122B-A10B13.4%estimated ± 4.2 pp, low confidence
175Nemotron 3 Ultra13.2%estimated ± 4.2 pp, low confidence
176GPT-512.9%estimated ± 4.6 pp, medium confidence
177Nemotron 3.5 Super VL 120B A12B11.8%estimated ± 4.2 pp, low confidence
178Nemotron 3.5 Super VL 120B A12B NVFP411.7%estimated ± 4.2 pp, low confidence
179MiniMax M2.711.6%estimated ± 4.3 pp, low confidence
180Laguna XS 2.111.4%estimated ± 4.2 pp, low confidence
181Grok Code Fast 111.3%estimated ± 4.2 pp, low confidence
182GPT-5.4 nano11.1%estimated ± 4.3 pp, low confidence
183Ornith-1.5-9B10.9%estimated ± 4.2 pp, low confidence
184Solar Pro 410.9%estimated ± 4.2 pp, low confidence
185Solar Open 210.5%estimated ± 4.2 pp, low confidence
186Jais-3-375B-A23B10.2%estimated ± 4.2 pp, low confidence
187Jais-3-7B10.2%estimated ± 4.2 pp, low confidence
188Grok 49.8%estimated ± 5.2 pp, low confidence
189Laguna XS.29.7%estimated ± 4.2 pp, low confidence
190Qwen3-Coder 480B-A35B9.1%estimated ± 4.2 pp, low confidence
191Ornith-1.0-9B8.8%estimated ± 4.2 pp, low confidence
192Qwen3.5-35B-A3B8.4%estimated ± 4.2 pp, low confidence
193K-EXAONE 2.06.7%estimated ± 4.2 pp, low confidence
194LongCat-Flash-Lite-Sparse6.7%estimated ± 4.2 pp, low confidence
195DeepSeek V3.12.8%estimated ± 4.2 pp, low confidence
196Kimi K22.4%estimated ± 4.2 pp, low confidence
197Ling 3.0 Flash VL0.1%estimated ± 6.6 pp, low confidence
198A.X K20.0%estimated ± 6.6 pp, low confidence
199Trinity-Large-Preview0.0%estimated ± 6.6 pp, low confidence
200Trinity-Large-Thinking0.0%estimated ± 6.6 pp, low confidence
201Nemotron 3.5 Bolt 74B-A9B0.0%estimated ± 6.6 pp, low confidence
202DeepSeek V3 03240.0%estimated ± 6.6 pp, low confidence
203GPT-OSS 20B0.0%estimated ± 6.6 pp, low confidence
204Mistral Small 40.0%estimated ± 6.6 pp, low confidence
205North Mini Code0.0%estimated ± 6.6 pp, low confidence
206Command A+0.0%estimated ± 6.6 pp, low confidence
207Mercury 2.50.0%estimated ± 6.6 pp, low confidence
208Gemini 2.5 Pro0.0%estimated ± 4.2 pp, low confidence

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General