mirobody/MetAgentBench
MetAgentBench MetAgentBench is an interactive clinical-decision benchmark for metabolic / GLP-1 agents: it scores actions, not essays. Each case is compiled from a published case report into a staged environment that answers only what is asked and charges for every action. Starting almost blind, the agent has to know what to ask, what to refuse, when to say it does not know, and whether to back down when the evidence contradicts it. The leaderboard reports five gated track… See the full description on the dataset page: https://huggingface.co/datasets/mirobody/MetAgentBench.
| Author / organization | mirobody |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 986 (as reported by Hugging Face) |
| Likes | 2 |
| Created | 7 October 2026 |
| Last updated at source | 8 October 2026 |
Popularity, last 2 days
986 now (+95 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
More from mirobody
- mirobody/ESL-BenchDataset
- mirobody/MedHall-BenchDataset
- mirobody/MedHarm-BenchDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.