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

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

#ModelScore
339Qwen1.5-4B-ChatQwen997
338gemma-2b-itGoogle1002
337gemma-1.1-2b-itGoogle1022
336PaLM 2Closed. Closed weights: ranked by its LMArena score, with no size or downloads to show.1027
335OLMo-7B-InstructAi21032
334Phi-3-mini-128k-instructMicrosoft1050
333Qwen-14B-ChatQwen1051
332Llama-2-7b-chat-hfMeta Llama1053
331Llama-3.2-1B-InstructMeta Llama1055
330gemma-7b-itGoogle1056
329granite-3.0-2b-instructIBM Granite1080
328Phi-3-mini-4k-instructMicrosoft1080
327SOLAR-10.7B-Instruct-v1.0Upstage1083
326Qwen1.5-7B-ChatQwen1084
325Llama-2-13b-chat-hfMeta Llama1084
324Mistral-7B-Instruct-v0.2Mistral AI1090
323gemma-1.1-7b-itGoogle1094
322GPT-3.5 Turbo (1106)Closed. Closed weights: ranked by its LMArena score, with no size or downloads to show.1094
321granite-3.0-8b-instructIBM Granite1096
320deepseek-llm-67b-chatDeepSeek1105
Show all 339 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.