Benchgen

BrowseComp — Results

RankModelScore
1nex-n2-5-max92.6
2kimi-k3-1-pro92.1
3gpt-6-astra91.5
4kimi-k391.2
5gpt-5-6-sol90.4
6nex-n2-5-pro89.7
7claude-fable-588
8gpt-5-6-terra87.5
9ornith-1-5-397b86.6
10seed-2-1-pro86.2
11gemini-3-1-pro85.9
12seed-2-1-turbo84.9
13claude-sonnet-584.7
14mistral-large-484.5
15gpt-5-584.4
16claude-opus-4-884.3
17hy384.2
18claude-opus-4-684
19minimax-m383.5
20deepseek-v4-pro83.4
21deepseek-v4-pro-max83.4
22gpt-5-6-luna83.3
23kimi-k2-683.2
24sage-a1-582.8
25gpt-5-482.7

BrowseComp — Evaluations

1 phaseActive

No evaluations yet for this benchmark.