Search on BrowseComp (test)
54.9AccuracyGPT-5-High
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-High# Samples=?, Training=?, Academic=No2026.05 | 54.9 | |
| GLM-4.7-357B# Samples=?, Training=?, Academic=No2026.05 | 52 | |
| OpenAI Deep Research# Samples=?, Training=?, Academic=No2026.05 | 51.5 | |
| DeepSeek-V3.2-671B# Samples=?, Training=?, Academic=No2026.05 | 51.4 | |
| OpenAI-o3# Samples=?, Training=?, Academic=No2026.05 | 49.1 | |
| OpenSeeker-v2-30B-SFT# Samples=10.6 k, Training=SFT, Academic=Yes2026.05 | 46 | |
| GLM-4.6-357B# Samples=?, Training=?, Academic=No2026.05 | 45.1 | |
| Minimax-M2-230B# Samples=?, Training=?, Academic=No2026.05 | 44 | |
| Tongyi DeepResearch# Samples=?, Training=CPT + SFT + RL, Academic=No2026.05 | 43.4 | |
| RedSearcher-30B# Samples=?, Training=CPT + SFT + RL, Academic=No2026.05 | 42.1 | |
| WebLeaper-30B-RL# Samples=?, Training=RL, Academic=No2026.05 | 38.8 | |
| Gemini-3-pro# Samples=?, Training=?, Academic=No2026.05 | 37.8 | |
| WebSailor-V2-30B-RL# Samples=?, Training=SFT + RL, Academic=No2026.05 | 35.3 | |
| DeepSeek-V3.1-671B# Samples=?, Training=?, Academic=No2026.05 | 30 | |
| OpenSeeker-v1-30B-SFT# Samples=11.7 k, Training=SFT, Academic=Yes2026.05 | 29.5 | |
| WebLeaper-30B-SFT# Samples=15 k, Training=SFT, Academic=No2026.05 | 27.7 | |
| WebSailor-V2-30B-SFT# Samples=?, Training=SFT, Academic=No2026.05 | 24.4 | |
| Claude-4.5-Sonnet# Samples=?, Training=?, Academic=No2026.05 | 24.1 | |
| Claude-4-Opus# Samples=?, Training=?, Academic=No2026.05 | 18.8 |