COMPL-AI
EvalsFramework / harness (not an eval — tooling/reference entry)
Maps EU AI Act requirements to technical benchmarks and scores models against them.
- Category
- Eval frameworks & harnesses
- Insurance exposure
- Regulatory/bias
- Publisher
- ETH Zurich / LatticeFlow / INSAIT
- Publisher type
- Academic
- License
- Apache-2.0
- Status
- Active (<90d)
- Last commit
- 2026-10-12
- Stars
- 213
- Notes
- Directly relevant to AI regulatory liability.
Wanich Risk Lab, 2026.