Model profile
GPT 5.5 Pro
Evidence summary
Safety. GPT 5.5 Pro has an estimated Safety rank of #32; its 90% source-sensitivity interval is #2–#259. Its behavior-only rank is #38; company governance moves the combined estimate to #32. Published Safety evidence spans 3 eval lineages and 3 of 7 components. Its strongest relative result is BioTIER (permit_compliance_pct, #2 of 52); its weakest is BullshitBench v2 (clear_pushback_rate, #47 of 114).
Freedom. GPT 5.5 Pro has an estimated Freedom rank of #187; its 90% source-sensitivity interval is #94–#389. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is Human Pathogen Capabilities Test (HPCT) — overall refusal (refusal_rate_pct, #1 of 69); its weakest is BioTIER (refuse_compliance_pct, #33 of 52).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-5-pro
- OpenRouter ↗openai/gpt-5.5-pro
- Official model documentation ↗Family-level model document · openai · first party
- Release source ↗declared identity
- openai.com ↗
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 |
|---|---|---|---|---|
| BioTIERpermit_compliance_pct | #2 / 52 | ↑99.9 | Source ↗official | |
| BioTIERrefuse_compliance_pct | #20 / 52 | ↑65.3 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #47 / 114 | ↑0.35 | Source ↗official | |
| Human Pathogen Capabilities Test (HPCT) — overall refusalrefusal_rate_pct | #27 / 69 | ↑0 | 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.