List your company

Pass@1 Evaluation on BBEH General Reasoning

9.94Pass@1OpenSIR
4.2725.74357.2158.6865Nov 1, 2025

Evaluation results

MethodLinks
OpenSIRBackbone=Llama-3.1-8B-Instruct2025.119.94
GRPOgsm8kBackbone=Llama-3.1-8B-Instruct2025.117.93
GRPOmathBackbone=Llama-3.1-8B-Instruct2025.117.79
OpenSIRBackbone=Llama-3.2-3B-Instruct2025.117.13
Absolute ZeroBackbone=Llama-3.1-8B-Instruct2025.116.76
R-ZeroBackbone=Llama-3.1-8B-Instruct2025.116.63
BaseBackbone=Llama-3.1-8B-Instruct2025.116.48
GRPOgsm8kBackbone=Llama-3.2-3B-Instruct2025.116.12
GRPOmathBackbone=Llama-3.2-3B-Instruct2025.115.98
R-ZeroBackbone=Llama-3.2-3B-Instruct2025.114.73
Absolute ZeroBackbone=Llama-3.2-3B-Instruct2025.114.54
BaseBackbone=Llama-3.2-3B-Instruct2025.114.49