List your company

Conversational on BS-Bench

70ScoreQwen3.5-9B
6.5623.0339.555.97May 29, 2026

Evaluation results

MethodLinks
Qwen3.5-9BParameters=9B, Variant=Thinking2026.0570
Qwen3.5-4BParameters=4B, Variant=Thinking2026.0563
OLMo-3-7BParameters=7B, Variant=Thinking2026.0523
Mellum 2 (RL)Post-training Stage=RL, Parameters=2.5B/12B, Variant=Thinking2026.0515
Mellum 2 (SFT)Post-training Stage=SFT, Parameters=2.5B/12B, Variant=Thinking2026.0514
Ministral-3-14BParameters=14B, Variant=Thinking2026.059