Model profile
Claude Instant 1.2
Evidence summary
Safety. Claude Instant 1.2 does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 3 of 7 components. Its strongest relative result is SORRY-Bench (avg, #5 of 51); its weakest is Claude 3 model-card adversarial human-preference evaluations (human_feedback_honesty_win_rate_pct, #4 of 4).
Freedom. Claude Instant 1.2 has an estimated Freedom rank of #629; its 90% source-sensitivity interval is #317–#655. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is Claude 3 model-card adversarial human-preference evaluations (human_feedback_harmlessness_win_rate_pct, #2 of 4); its weakest is SORRY-Bench (avg, #45 of 51).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Official model documentation ↗Family-level model document · anthropic · first party
- Release source ↗direct research preview identity date
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Claude 3 model-card adversarial human-preference evaluationsdiscrimination_rank | #5 / 5 | ↓5 | Source ↗official | |
| Claude 3 model-card adversarial human-preference evaluationshuman_feedback_harmlessness_win_rate_pct | #3 / 4 | ↑85 | Source ↗official | |
| Claude 3 model-card adversarial human-preference evaluationshuman_feedback_honesty_win_rate_pct | #4 / 4 | ↑61 | Source ↗official | |
| SORRY-Benchavg | #5 / 51 | ↓0.08 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Claude 3 model-card adversarial human-preference evaluationshuman_feedback_harmlessness_win_rate_pct | #2 / 4 | ↓85 | Source ↗official | |
| SORRY-Benchavg | #45 / 51 | ↑0.08 | Source ↗official |