List your company

Mathematical Reasoning on GSM-Hard OOD 1.0 (test)

12Greedy Success RateRiM
3.165.4557.7510.045May 28, 2026

Evaluation results

MethodLinks
RiMModel=Llama-3.2-3B, Variant=Final block, TTFT (ms)=27.92026.051214.1
RiMModel=Llama-3.2-1B, Variant=Final block, TTFT (ms)=16.12026.0510.513.8
CoconutModel=Llama-3.2-3B, Variant=w/ Stage 0, TTFT (ms)=188.82026.0510.213.5
CoconutModel=Llama-3.2-1B, Variant=w/ Stage 0, TTFT (ms)=108.32026.058.512.2
SFTModel=Llama-3.2-3B, Variant=w/o CoT, TTFT (ms)=27.92026.058.510.8
RiMModel=GPT-2, Variant=Final block, TTFT (ms)=7.62026.057.811.2
CoconutModel=GPT-2, Variant=w/ Stage 0, TTFT (ms)=53.42026.057.110.7
SFTModel=Llama-3.2-1B, Variant=w/o CoT, TTFT (ms)=16.12026.055.39.5
SFTModel=GPT-2, Variant=w/o CoT, TTFT (ms)=7.62026.053.57.6