List your company

Reasoning on GPQA Diamond (pass@4 accuracy)

67.17Pass@4 AccuracySAGC
54.565257.837661.1164.3824Jun 1, 2026

Evaluation results

MethodLinks
SAGCBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=Dynamic2026.0667.17
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=162026.0663.64
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=42026.0662.63
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=82026.0662.63
SAGCBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=Dynamic2026.0662.12
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=42026.0661.11
SAGCBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=Dynamic2026.0661.11
OpenRLHFBackbone=Llama3.2-3b, Algorithm=DAPO, Group Size=162026.0661.11
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=162026.0660.61
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=82026.0657.58
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=GRPO, Group Size=82026.0657.07
OpenRLHFBackbone=Llama3.2-3b, Algorithm=GRPO, Group Size=42026.0657.07
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=82026.0656.06
SAGCBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=Dynamic2026.0656.06
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=42026.0655.05
OpenRLHFBackbone=Qwen2.5-3b, Algorithm=DAPO, Group Size=162026.0655.05