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

141 models in math ranked by LMArena score · updated Nov 11, 2026

#ModelScore
141gemma-2b-itGoogle1009
140OLMo-7B-InstructAi21018
139Qwen1.5-4B-ChatQwen1026
138Llama-2-7b-chat-hfMeta Llama1042
137gemma-1.1-2b-itGoogle1047
136SOLAR-10.7B-Instruct-v1.0Upstage1064
135Llama-2-13b-chat-hfMeta Llama1065
134gemma-7b-itGoogle1066
133Qwen-14B-ChatQwen1068
132Qwen1.5-7B-ChatQwen1080
131Mistral-7B-Instruct-v0.2Mistral AI1085
130Llama-3.2-1B-InstructMeta Llama1086
129Phi-3-mini-128k-instructMicrosoft1089
128Llama-2-70b-chat-hfMeta Llama1091
127gemma-1.1-7b-itGoogle1107
126deepseek-llm-67b-chatDeepSeek1108
125granite-3.0-2b-instructIBM Granite1117
124Qwen1.5-14B-ChatQwen1125
123Llama-3.2-3B-InstructMeta Llama1126
122gemma-2-2b-itGoogle1135
Show all 141 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.