List your company

Reasoning on GPQA Diamond (Avg@5, Weighted Average)

56.67Avg@5DelTA
42.900446.475250.0553.6248May 20, 2026

Evaluation results

MethodLinks
DelTABackbone=Qwen3-14B-Base2026.0556.6768.4
DAPOBackbone=Qwen3-14B-Base2026.0554.5566.77
DelTABackbone=Qwen3-8B-Base2026.055062.38
DAPOBackbone=Qwen3-8B-Base2026.0543.4358.87