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

147 models in creative writing ranked by LMArena score · updated Nov 11, 2026

#ModelScore
147Qwen1.5-4B-ChatQwen969
146gemma-2b-itGoogle986
145OLMo-7B-InstructAi2990
144gemma-1.1-2b-itGoogle998
143gemma-7b-itGoogle1023
142Phi-3-mini-128k-instructMicrosoft1024
141Qwen-14B-ChatQwen1028
140Llama-3.2-1B-InstructMeta Llama1033
139Llama-2-7b-chat-hfMeta Llama1033
138Qwen1.5-7B-ChatQwen1034
137Phi-3-mini-4k-instructMicrosoft1044
136granite-3.0-2b-instructIBM Granite1046
135Llama-2-13b-chat-hfMeta Llama1047
134gemma-1.1-7b-itGoogle1059
133deepseek-llm-67b-chatDeepSeek1067
132Mistral-7B-Instruct-v0.2Mistral AI1068
131granite-3.0-8b-instructIBM Granite1071
130Llama-2-70b-chat-hfMeta Llama1075
129Phi-3-small-8k-instructMicrosoft1083
128Qwen1.5-32B-ChatQwen1083
Show all 147 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.