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

#ModelScore
339gemma-2b-itGoogle970
338OLMo-7B-InstructAi2978
337Qwen1.5-4B-ChatQwen978
336gemma-1.1-2b-itGoogle992
335Llama-2-7b-chat-hfMeta Llama1006
334PaLM 2Closed. Closed weights: ranked by its LMArena score, with no size or downloads to show.1010
333gemma-7b-itGoogle1017
332Phi-3-mini-128k-instructMicrosoft1023
331Llama-3.2-1B-InstructMeta Llama1031
330Qwen-14B-ChatQwen1031
329Llama-2-13b-chat-hfMeta Llama1045
328SOLAR-10.7B-Instruct-v1.0Upstage1046
327granite-3.0-2b-instructIBM Granite1055
326gemma-1.1-7b-itGoogle1057
325Phi-3-mini-4k-instructMicrosoft1058
324Qwen1.5-7B-ChatQwen1058
323Mistral-7B-Instruct-v0.2Mistral AI1060
322Llama-2-70b-chat-hfMeta Llama1071
321deepseek-llm-67b-chatDeepSeek1079
320Phi-3-small-8k-instructMicrosoft1086
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.