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 in instructions ranked by LMArena score · updated Nov 11, 2026

#ModelScore
148gemma-2b-itGoogle970
147OLMo-7B-InstructAi2978
146Qwen1.5-4B-ChatQwen978
145gemma-1.1-2b-itGoogle992
144Llama-2-7b-chat-hfMeta Llama1006
143gemma-7b-itGoogle1017
142Phi-3-mini-128k-instructMicrosoft1023
141Llama-3.2-1B-InstructMeta Llama1031
140Qwen-14B-ChatQwen1031
139Llama-2-13b-chat-hfMeta Llama1045
138SOLAR-10.7B-Instruct-v1.0Upstage1046
137granite-3.0-2b-instructIBM Granite1055
136gemma-1.1-7b-itGoogle1057
135Phi-3-mini-4k-instructMicrosoft1058
134Qwen1.5-7B-ChatQwen1058
133Mistral-7B-Instruct-v0.2Mistral AI1060
132Llama-2-70b-chat-hfMeta Llama1071
131deepseek-llm-67b-chatDeepSeek1079
130Phi-3-small-8k-instructMicrosoft1086
129granite-3.0-8b-instructIBM Granite1088
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.