Model profile
Claude 2.1
Anthropicdeveloper
2023-11-21release date
#137 / 267overall rank
3eval lineages
Evidence summary
Claude 2.1 has an estimated overall rank of #137; its 90% source-sensitivity interval is #1–#232. Its behavior-only rank is #153; company governance moves the combined estimate to #137. Published evidence spans 3 evals and 2 of 7 behavior components. Its strongest relative result is SORRY-Bench (avg, #1 of 51); its weakest is OR-Bench (over_refusal_rate, #25 of 25).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- OpenRouter ↗anthropic/claude-2.1
- 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.