List your company

Multi-hop Numeric Reasoning QA on HotpotQA Dynamic Numeric Reasoning suite

2,252Tokens (No-adv)Qwen-Plus
307.8864812.60821,317.331,822.0518Jan 19, 2026

Evaluation results

MethodLinks
Qwen-PlusModel=Qwen-plus2026.012,25215455,665.3524.727898.592.58043.451
DS V32Model=DeepSeek-V3.22026.011,548.6815710,720.526.9272967863.520.702
Gemini 2.5 FlashModel=Gemini-2.5-Flash2026.01940.681509,795.0310.416298.5887121.749
GPT-5.1Model=GPT-5.12026.01447.291723,375.657.558197.5845113.211
DS R1Model=DeepSeek-R1-05282026.01382.661507,995.6420.797510098.9993.9724.805