BrowseComp-zh

BrowseComp-zh - Chinese web browsing and comprehension agentic benchmark.

Category

General Agents

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for BrowseComp-zh

Model Score Date Verified
Qwen3.5-122B-A10B
100.00%
01.02.2026 Verified
Qwen3.5-27B
88.29%
01.02.2026 Unverified
Qwen3.5-35B-A3B
99.40%
01.02.2026 Unverified
GPT-5-mini 2025-08-07
69.37%
01.02.2026 Unverified
MiniCPM5-2B
60.36%
— Verified
LFM2.5-2.6B
9.76%
— Verified
Gemma4-E2B
2.10%
— Verified
Qwen3.5-4B
54.50%
— Verified
granite-4.2-3B
26.73%
— Verified
Nemotron-3-Nano-4B
3.30
— Verified
Gemma4-E4B
5.56%
— Verified
LFM2.5-8B-A1B
14.86%
— Verified
Qwen3.5-2B
22.37%
— Verified