Model profile
GPT 5.5 Instant
Evidence summary
Safety. GPT 5.5 Instant has an estimated Safety rank of #12; its 90% source-sensitivity interval is #3–#87. Its behavior-only rank is #17; company governance moves the combined estimate to #12. Published Safety evidence spans 4 eval lineages and 3 of 7 components. Its strongest relative result is Arena Factuality — Text Arena (factuality-only weighting) (factuality_bt_rating, #9 of 111); its weakest is SM-Bench (eq_boundaries, #50 of 84).
Freedom. GPT 5.5 Instant has an estimated Freedom rank of #598; its 90% source-sensitivity interval is #332–#642. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (overfit, #36 of 84); its weakest is SM-Bench (adversarial, #77 of 84).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-5-instant-05-26
- Official model documentation ↗Family-level model document · openai · first party
- Release source ↗declared identity
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 |
|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #110 / 345 | ↓0.653 | Source ↗official | |
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #9 / 111 | ↑1473.0 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #52 / 117 | ↑0.34 | Source ↗official | |
| SM-Benchadversarial | #8 / 84 | ↑88.78 | Source ↗official | |
| SM-Benchambiguous_interpretation | #30 / 84 | ↑88.1 | Source ↗official | |
| SM-Benchanti_hallucination | #34 / 84 | ↑95.81 | Source ↗official | |
| SM-Bencheq_boundaries | #50 / 84 | ↑62.36 | Source ↗official | |
| SM-Benchoverfit | #36 / 84 | ↑78.96 | 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.