Medical Reasoning on Healthbench Hard (Pass, BoN, FewTool, SolAgg, SummAgg, AggAgent)
40.91Pass RatePass
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| PassK=8, Model=GLM-4.7-Flash2026.04 | 40.91 | — | — | — | — | — | — | — | |
| PassK=4, Model=GLM-4.7-Flash2026.04 | 31.96 | — | — | — | — | — | — | — | |
| PassK=2, Model=GLM-4.7-Flash2026.04 | 21.34 | — | — | — | — | — | — | — | |
| PassK=1, Model=GLM-4.7-Flash2026.04 | 8.67 | — | — | — | — | — | — | — | |
| AggAgentK=2, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | — | 24.45 | |
| AggAgentK=4, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | — | 25.36 | |
| AggAgentK=8, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | — | 27.99 | |
| BoNK=2, Model=GLM-4.7-Flash2026.04 | — | — | — | 9.8 | — | — | — | — | |
| BoNK=4, Model=GLM-4.7-Flash2026.04 | — | — | — | 9.65 | — | — | — | — | |
| BoNK=8, Model=GLM-4.7-Flash2026.04 | — | — | — | 9.91 | — | — | — | — | |
| FewToolK=2, Model=GLM-4.7-Flash2026.04 | — | — | — | — | 9.32 | — | — | — | |
| FewToolK=4, Model=GLM-4.7-Flash2026.04 | — | — | — | — | 8.48 | — | — | — | |
| FewToolK=8, Model=GLM-4.7-Flash2026.04 | — | — | — | — | 8.9 | — | — | — | |
| SolAggK=2, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | 10.5 | — | — | |
| SolAggK=4, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | 13.7 | — | — | |
| SolAggK=8, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | 15.72 | — | — | |
| SummAggK=2, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | 1.21 | — | |
| SummAggK=4, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | 4.76 | — | |
| SummAggK=8, Model=GLM-4.7-Flash2026.04 | — | — | — | — | — | — | 7.35 | — |