Benchgen

BrowseComp-zh — Results

RankModelScore
1qwen3-5-397b-a17b0.703
2qwen3-5-122b-a10b0.699
3qwen3-5-35b-a3b0.695
4longcat-flash-thinking0.69
5glm-4-70.666
6deepseek-v3-2-thinking0.65
7deepseek-v3-20.65
8kimi-k2-thinking-09050.623
9qwen3-5-27b0.621
10deepseek-v3-10.492
11minimax-m20.485
12deepseek-v3-2-exp0.479
13deepseek-r1-05280.357
B

BrowseComp-zh — Phases

1 phaseActive

Phase 1·

BrowseComp-zh

Jul 24, 2026
Current