List your company

Math Reasoning on MATH (Pass@8)

98.4Pass@8RFT
95.90496.55297.297.848Apr 13, 2026

Evaluation results

MethodLinks
RFTBase Model=Qwen3-4B-Instruct2026.0498.4
GRPOBase Model=Qwen3-4B-Instruct2026.0498
Olmo-3-7B-InstructBase Model=Olmo-3-7B-Instruct2026.0498
SRTBase Model=Olmo-3-7B-Instruct2026.0498
Qwen3-4B-InstructBase Model=Qwen3-4B-Instruct2026.0497.6
SRTBase Model=Qwen3-4B-Instruct2026.0497.6
SFTBase Model=Qwen3-4B-Instruct2026.0497
SDFTBase Model=Qwen3-4B-Instruct2026.0497
SD-ZEROBase Model=Qwen3-4B-Instruct2026.0497
RFTBase Model=Olmo-3-7B-Instruct2026.0497
GRPOBase Model=Olmo-3-7B-Instruct2026.0497
SDFTBase Model=Olmo-3-7B-Instruct2026.0497
SD-ZEROBase Model=Olmo-3-7B-Instruct2026.0497
SFTBase Model=Olmo-3-7B-Instruct2026.0496