Medical Deep Search on QuarkMedSearch Benchmark
58.57ScoreGLM-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-5Backbone=GLM-5, Size=744B-A40B2026.04 | 58.57 | |
| DeepSeek-V3.2Backbone=DeepSeek-V3.2, Size=671B-A37B2026.04 | 57.14 | |
| Kimi-K2.5Backbone=Kimi-K2.5, Size=1T-A32B2026.04 | 56.74 | |
| Seed1.8Backbone=Seed1.8, Size=-2026.04 | 55.71 | |
| QuarkMedSearchBackbone=QuarkMedSearch, Size=30B-A3B2026.04 | 55.71 | |
| Qwen3.5-PlusBackbone=Qwen3.5-Plus, Size=397B-A17B2026.04 | 51.42 | |
| Qwen3.5-27BBackbone=Qwen3.5-27B, Size=27B2026.04 | 44.28 | |
| Qwen3.5-35B-A3BBackbone=Qwen3.5-35B-A3B, Size=35B-A3B2026.04 | 43.57 | |
| Tongyi DeepResearchBackbone=Tongyi DeepResearch, Size=30B-A3B2026.04 | 40.71 |