List your company

Expert-level Science Question Answering on GPQA Diamond

78Scoregpt-oss-120b
24.190438.160252.1366.0998Jan 11, 2026Jan 27, 2026Feb 13, 2026Mar 2, 2026Mar 18, 2026Apr 4, 2026Apr 21, 2026

Evaluation results

MethodLinks
gpt-oss-120bNumber of Parameters=117B, Optimization Level=high2026.0178
GLM-4.5-AirNumber of Parameters=110B2026.0175.8
gpt-oss-120bNumber of Parameters=117B, Optimization Level=medium2026.0169.4
Solar OpenNumber of Parameters=102B2026.0168.1
BaseModel=Qwen-2.5-7B-Inst, Evaluation Protocol=0-shot2026.0436.36
POPModel=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0435.35
BaseModel=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0433.84
POPModel=Qwen-2.5-7B-Inst, Evaluation Protocol=0-shot2026.0433.84
Train on DModel=Qwen-2.5-7B, Evaluation Protocol=0-shot2026.0428.28
Train on DModel=Qwen-2.5-7B-Inst, Evaluation Protocol=0-shot2026.0426.26