Model profile
Claude 1
2023-03-14release date
2eval lineages
Evidence summary
Published evidence spans 2 evals and 1 of 7 behavior components. Its strongest relative result is HarmBench (dr, #6 of 28); its weakest is Do-Not-Answer (human_harmlessness_rate, #3 of 6).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- OpenRouter ↗anthropic/claude-1
- Official model documentation ↗Family-level model document · anthropic · first party
- Release source ↗direct research preview identity date
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| Do-Not-Answerhuman_harmlessness_rate | #3 / 6 | 98.3 | ↑ higher | Source ↗official | |
| HarmBenchdr | #6 / 28 | 5 | ↓ lower | Source ↗official |