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

148 models ranked by LMArena score · updated Nov 11, 2026

#ModelScore
148Qwen1.5-4B-ChatQwen997
147gemma-2b-itGoogle1002
146gemma-1.1-2b-itGoogle1022
145OLMo-7B-InstructAi21032
144Phi-3-mini-128k-instructMicrosoft1050
143Qwen-14B-ChatQwen1051
142Llama-2-7b-chat-hfMeta Llama1053
141Llama-3.2-1B-InstructMeta Llama1055
140gemma-7b-itGoogle1056
139granite-3.0-2b-instructIBM Granite1080
138Phi-3-mini-4k-instructMicrosoft1080
137SOLAR-10.7B-Instruct-v1.0Upstage1083
136Qwen1.5-7B-ChatQwen1084
135Llama-2-13b-chat-hfMeta Llama1084
134Mistral-7B-Instruct-v0.2Mistral AI1090
133gemma-1.1-7b-itGoogle1094
132granite-3.0-8b-instructIBM Granite1096
131deepseek-llm-67b-chatDeepSeek1105
130Llama-3.2-3B-InstructMeta Llama1110
129Phi-3-small-8k-instructMicrosoft1110
Show all 148 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.