List your company

Multi-turn Conversation Evaluation on IMDb

5.6MT-Bench ScoreCOALA
0.8162.0583.34.542May 22, 2026

Evaluation results

MethodLinks
COALABase Model=DOLPHIN-7B2026.055.6
ORPOBase Model=DOLPHIN-7B2026.055.5
DPOBase Model=DOLPHIN-7B2026.055.1
SFTBase Model=LLAMA-8B2026.053.5
COALABase Model=LLAMA-8B2026.053.5
SFTBase Model=DOLPHIN-7B2026.053.2
DPOBase Model=LLAMA-8B2026.053.1
ORPOBase Model=LLAMA-8B2026.052.9
COALABase Model=MISTRAL-7B2026.052.9
DPOBase Model=MISTRAL-7B2026.052.8
SFTBase Model=MISTRAL-7B2026.052.3
ORPOBase Model=MISTRAL-7B2026.051.7
DPOBase Model=DISTILGPT2026.051.6
SFTBase Model=GPT-22026.051.4
SFTBase Model=DISTILGPT2026.051.2
ORPOBase Model=DISTILGPT2026.051.2
COALABase Model=DISTILGPT2026.051.2
ORPOBase Model=GPT-22026.051.1
COALABase Model=GPT-22026.051.1
DPOBase Model=GPT-22026.051