Model profile
GPT 5.4 Nano
OpenAIdeveloper
2026-03-17release date
#197 / 267overall rank
12eval lineages
Evidence summary
GPT 5.4 Nano has an estimated overall rank of #197; its 90% source-sensitivity interval is #85–#235. Its behavior-only rank is #209; company governance moves the combined estimate to #197. Published evidence spans 12 evals and 5 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #53 of 311); its weakest is CAIS Risk Index (political_manipulation, #32 of 32).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-4-nano
- OpenRouter ↗openai/gpt-5.4-nano
- Official model documentation ↗Family-level model document · openai · 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 |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #53 / 311 | 0.504 | ↓ lower | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #80 / 105 | 0.1233 | ↑ higher | Source ↗official | |
| CAIS Risk Indexagent_red_teaming | #24 / 43 | 81.7 | ↓ lower | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #31 / 48 | 78.3 | ↓ lower | Source ↗official | |
| CAIS Risk Indexhle_overconfidence | #16 / 49 | 47.1 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmachiavelli | #24 / 45 | 88.5 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmask | #29 / 51 | 23.7 | ↓ lower | Source ↗official | |
| CAIS Risk Indexpolitical_manipulation | #32 / 32 | 63.3 | ↓ lower | Source ↗official | |
| CAIS Risk Indextextquests_harm | #10 / 48 | 15.2 | ↓ lower | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #24 / 105 | 9.521 | ↓ lower | Source ↗official | |
| MACHIAVELLIdeception_relative_random_pct | #24 / 50 | 88.5 | ↓ lower | Source ↗official | |
| SM-Benchadversarial | #73 / 73 | 68.29 | ↑ higher | Source ↗official | |
| SM-Benchambiguous_interpretation | #31 / 73 | 86.31 | ↑ higher | Source ↗official | |
| SM-Benchanti_hallucination | #70 / 73 | 68.06 | ↑ higher | Source ↗official | |
| SM-Bencheq_boundaries | #47 / 73 | 58.71 | ↑ higher | Source ↗official | |
| SM-Benchoverfit | #73 / 73 | 4.37 | ↑ higher | Source ↗official | |
| VETO Misfired Alignmentmisfired_alignment_rate_pct | #25 / 25 | 18.9 | ↓ lower | Source ↗official |