List your company

General Reasoning on GPQA (Acc, Cost)

48.13AccuracyCoT
1.340413.487725.63537.7823Mar 14, 2026

Evaluation results

MethodLinks
CoTModel=Gemma32026.0348.134,926
CoTModel=Qwen32026.0344.84,089
CoTModel=Llama3.12026.0344.064,156
DSTModel=Gemma32026.035.374,940
DSTModel=Qwen32026.034.94,141
DSTModel=Llama3.12026.034.483,994
ToTModel=Gemma32026.034.110,602
ToTModel=Qwen32026.033.768,875
ToTModel=Llama3.12026.033.758,678
DPTSModel=Gemma32026.033.448,336
DPTSModel=Qwen32026.033.277,001
DPTSModel=Llama3.12026.033.146,892