List your company

Math Reasoning on HMMT25 (Pass@8)

66.7Pass@8SRT
45.951.356.762.1Apr 13, 2026

Evaluation results

MethodLinks
SRTBase Model=Olmo-3-7B-Instruct2026.0466.7
SD-ZEROBase Model=Olmo-3-7B-Instruct2026.0466.7
SD-ZEROBase Model=Qwen3-4B-Instruct2026.0463.3
SRTBase Model=Qwen3-4B-Instruct2026.0460
RFTBase Model=Olmo-3-7B-Instruct2026.0460
SDFTBase Model=Olmo-3-7B-Instruct2026.0456.7
RFTBase Model=Qwen3-4B-Instruct2026.0453.3
Qwen3-4B-InstructBase Model=Qwen3-4B-Instruct2026.0450
SDFTBase Model=Qwen3-4B-Instruct2026.0450
SFTBase Model=Qwen3-4B-Instruct2026.0446.7
GRPOBase Model=Qwen3-4B-Instruct2026.0446.7
Olmo-3-7B-InstructBase Model=Olmo-3-7B-Instruct2026.0446.7
SFTBase Model=Olmo-3-7B-Instruct2026.0446.7
GRPOBase Model=Olmo-3-7B-Instruct2026.0446.7