Web Browsing Competition (Chinese) on Browse Comp ZH
81.3ScoreSeed-1.8
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed-1.8Model Access Type=Closed-Source2026.02 | 81.3 | |
| DeepSeek-V3.2Model Access Type=Closed-Source2026.02 | 65 | |
| OpenAI-GPT-5-highModel Access Type=Closed-Source2026.02 | 65 | |
| GLM-4.6Model Access Type=Closed-Source2026.02 | 49.5 | |
| Minimax-M2Model Access Type=Closed-Source2026.02 | 48.5 | |
| Tongyi-DeepResearch 30BModel Access Type=Open-Source2026.02 | 46.7 | |
| IterResearch-30B-A3BModel Access Type=Open-Source2026.02 | 45.2 | |
| WebSailor-V2-30B-A3BModel Access Type=Open-Source2026.02 | 44.1 | |
| Claude-4.5-SonnetModel Access Type=Closed-Source2026.02 | 40.8 | |
| MiroThinker 8BModel Access Type=Open-Source2026.02 | 40.2 | |
| AgentCPM-Explore-4BModel Access Type=Open-Source2026.02 | 29.1 | |
| SFT + ACTGUIDE-RL2026.05 | 28.02 | |
| Merged-Model-4BModel Access Type=Open-Source2026.02 | 26.8 | |
| SFT + RL2026.05 | 26.3 | |
| SFT2026.05 | 25.61 | |
| ZeroRL + ACTGUIDE-RL2026.05 | 20.41 | |
| WebDancer-QwQ-32BModel Access Type=Open-Source2026.02 | 18 | |
| ZeroRL2026.05 | 15.26 |