Through an API

Open and closed models answer the same suite-standard 1.0 through their APIs, ranked together. 1 ranked model.

#ModelPass rate
1DeepSeek V4.1 Flash 1017Open weights100%

Model leaderboard

86 models in expert ranked by LMArena score · updated Nov 11, 2026

#ModelScore
86Llama-3.2-1B-InstructMeta Llama1007
85granite-3.0-2b-instructIBM Granite1064
84granite-3.0-8b-instructIBM Granite1087
83Llama-3.2-3B-InstructMeta Llama1090
82Mixtral-8x22B-Instruct-v0.1Mistral AI1113
81granite-3.1-2b-instructIBM Granite1131
80QwQ-32B-PreviewQwen1134
79granite-3.1-8b-instructIBM Granite1141
78Llama-3.1-8B-InstructMeta Llama1144
77Llama-3.1-70B-InstructMeta Llama1208
76Qwen2.5-Coder-32B-InstructQwen1220
75Llama-3.3-70B-InstructMeta Llama1225
74Llama-4-Scout-17B-16E-InstructMeta Llama1235
73Llama-3.1-405B-InstructMeta Llama1243
72Qwen2.5-72B-InstructQwen1245
71granite-4.0-h-smallIBM Granite1251
70Llama-4-Maverick-17B-128E-InstructMeta Llama1258
69gpt-oss-20bOpenAI1259
68granite-4.2-3bIBM Granite1305
67Olmo-3.1-32B-InstructAi21306
Show all 86 models

Score by size

Plot against
Best for its size

Scores: LMArena, CC BY 4.0. Measured on served versions, not local builds. A model with several served variants shows its best.