HarmBench vs Inspect AI
| HarmBench | Inspect AI | |
|---|---|---|
| Free plan | No | Yes |
| Free trial | No | No |
| Paid from | — | Free |
| Open source | No | No |
| Platforms | Linux | api, extension |
| Free plan | Yes | — |
| Deployment options | self-hosted | self-hosted |
| LLM-as-a-judge | Yes | Yes |
| Safety evaluations | Yes | Yes |
| Custom metrics | — | Yes |
| Human review workflows | — | Yes |
| CI/CD integration | — | Yes |
Both are listed in Best LLM Evaluation Tools. On Everything Xiaomi, Inspect AI scores higher on our published basis.