List your company

Conversation Simulation on MirrorBench

68.3Primary MetricOSIM 8B
11.72426.41241.155.788Jun 12, 2026

Evaluation results

MethodLinks
OSIM 8BModel=OSIM, Parameter Count=8B, Training Stage=Final2026.0668.3
Claude Opus 4.72026.0663.7
GPT 5.52026.0656.7
Qwen3 8B InstModel=Qwen3-8B-Instruct, Parameter Count=8B2026.0654
OSIM 8B-MidModel=OSIM, Parameter Count=8B, Training Stage=Mid-trained2026.0649.1
Gemini 3.1 Pro2026.0648.3
Qwen 3.6 Plus2026.0648
Others *2026.0645.4
Base 8BModel=Qwen3-8B-Base, Parameter Count=8B2026.0613.9