Share your thoughts, 1 month free Claude Pro on usSee more

Multi-turn Conversation on MT-Bench (AVG Metric)

85.25Average Score

CoDIT-Qwen3-8B

Updated 3d ago

Evaluation Results

Method	Links
CoDIT-Qwen3-8B 2026.04		85.25
CoDIT-Qwen3-30B 2026.04		84.18
CoDIT-Gemma3 2026.04		83.13
WebR-Pro 2026.04		82.72
Llama-3.1-LMSYS-Chat-1M-Synth 2026.04		80.05
WebR-Basic 2026.04		75.43
CoDIT-Qwen3-8B 2026.04		75.39
Gemma-2-LMSYS-Chat-1M-Synth 2026.04		74.56
CoDIT-Gemma3 2026.04		74.33
Magpie-Pro-300K-Filtered 2026.04		74.05
CoDIT-Qwen3-30B 2026.04		73.73
Llama-3.1-LMSYS-Chat-1M-Synth 2026.04		72.11
WildChat 2026.04		70.04
WebR-Pro 2026.04		70
Gemma-2-LMSYS-Chat-1M-Synth 2026.04		69.51
WebR-Basic 2026.04		64.31
Magpie-Pro-300K-Filtered 2026.04		63.15
WildChat 2026.04		62.24