asha-benchmarks
OtherGitHubActivePython
Asha, DNAi's medical AI agent: the same language model alone and inside Asha, same items, same grader. Per-item, re-gradable artifacts for HealthBench (5,000, Hard, Professional, Vishwanath 500), MedQA, psychosis-bench, enclomiphene-bench, a locked pediatric chart, plus model cards and A2A agent cards.
Description provenance: Original source
Measured: Oct 18, 2026 · 6 observations
Domains & Uses
- Domains
- Agents, skills and MCP
- Uses
- Observability
Check at the source
Compatibility and prerequisites must be checked in the official documentation.
https://github.com/EndlessRay/asha-benchmarksClassifications
- Nature
- Tool