Deep Search on HLE text-only
40.8ScoreDeepSeek-V3.2-671B
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepSeek-V3.2-671BCategory=Large Foundation Model2026.02 | 40.8 | |
| Tongyi-DeepResearch-30BCategory=Research Agent2026.02 | 32.9 | |
| Minimax-M2-230BCategory=Large Foundation Model2026.02 | 31.8 | |
| GLM-4.6-357BCategory=Large Foundation Model2026.02 | 30.4 | |
| Nanbeige4.1-3BCategory=Ours2026.02 | 22.29 | |
| MiroThinker-v1.0-8BCategory=Research Agent2026.02 | 21.5 | |
| AgentCPM-Explore-4BCategory=Research Agent2026.02 | 19.1 | |
| Qwen3-30B-A3B-2507Category=Small Foundation Model2026.02 | 14.81 | |
| Nanbeige4-3B-2511Category=Baseline2026.02 | 13.89 | |
| Qwen3-4B-2507Category=Small Foundation Model2026.02 | 11.13 | |
| Qwen3-8BCategory=Small Foundation Model2026.02 | 10.24 | |
| Qwen3-14BCategory=Small Foundation Model2026.02 | 10.17 | |
| Qwen3-32BCategory=Small Foundation Model2026.02 | 9.26 | |
| Qwen3-Next-80B-A3BCategory=Small Foundation Model2026.02 | 9.26 |