List your company

Math Reasoning on AIME 25 (Pass@8)

80Pass@8RFT
66.16869.75973.3576.941Apr 13, 2026

Evaluation results

MethodLinks
RFTBase Model=Qwen3-4B-Instruct2026.0480
SRTBase Model=Qwen3-4B-Instruct2026.0480
SD-ZEROBase Model=Qwen3-4B-Instruct2026.0480
SD-ZEROBase Model=Olmo-3-7B-Instruct2026.0480
Qwen3-4B-InstructBase Model=Qwen3-4B-Instruct2026.0476.7
SRTBase Model=Olmo-3-7B-Instruct2026.0476.7
SFTBase Model=Qwen3-4B-Instruct2026.0473.3
Olmo-3-7B-InstructBase Model=Olmo-3-7B-Instruct2026.0473.3
RFTBase Model=Olmo-3-7B-Instruct2026.0473.3
SDFTBase Model=Olmo-3-7B-Instruct2026.0473.3
GRPOBase Model=Qwen3-4B-Instruct2026.0470
SDFTBase Model=Qwen3-4B-Instruct2026.0470
SFTBase Model=Olmo-3-7B-Instruct2026.0466.7
GRPOBase Model=Olmo-3-7B-Instruct2026.0466.7