List your company

Role-playing evaluation on DynSess-Eval

4.35Average Performance (Auto)Claude Sonnet 4.6
2.81083.21043.614.0096May 28, 2026

Evaluation results

MethodLinks
Claude Sonnet 4.6Domain=General Models2026.054.353.243.763.124.553.214.883.374.243.27
GPT-5.4Domain=General Models2026.054.293.313.83.234.273.344.883.384.293.28
Deepseek v3.2Domain=General Models, Params=685B2026.054.23.213.73.114.253.174.733.344.123.22
Gemini-3-proDomain=General Models2026.054.123.173.763.053.673.064.833.364.233.2
Doubao-1.5-pro-characterDomain=Character Models2026.053.963.383.83.293.663.424.53.473.973.35
Qwen-plus-characterDomain=Character Models2026.053.663.013.572.863.462.884.023.223.593.06
MiniMax-M2-HERDomain=Character Models, Params=32B2026.053.232.993.182.952.892.913.763.133.112.97
CoserDomain=Character Models, Params=70B2026.052.872.962.612.892.372.93.343.093.122.98