Model profile
GPT 5 Nano
OpenAIdeveloper
2025-08-07release date
#11 / 267overall rank
20eval lineages
Evidence summary
GPT 5 Nano has an estimated overall rank of #11; its 90% source-sensitivity interval is #5–#55. Its behavior-only rank is #8; company governance moves the combined estimate to #11. Published evidence spans 20 evals and 7 of 7 behavior components. Its strongest relative result is HELM Safety (simple_safety_tests, #1 of 80); its weakest is CAIS Risk Index (hle_overconfidence, #44 of 49).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-nano
- OpenRouter ↗openai/gpt-5-nano
- System card ↗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 | #48 / 311 | 0.4885 | ↓ lower | Source ↗official | |
| AIRBench 2024 Safety Scenariossafety_scenarios | #6 / 80 | 0.878 | ↑ higher | Source ↗official | |
| ANIMAscore | #9 / 19 | 0.708 | ↑ higher | Source ↗official | |
| CAIS Risk Indexagent_red_teaming | #31 / 43 | 88.5 | ↓ lower | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #27 / 48 | 71.3 | ↓ lower | Source ↗official | |
| CAIS Risk Indexhle_overconfidence | #44 / 49 | 80 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmachiavelli | #1 / 45 | 79.4 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmask | #16 / 51 | 12.3 | ↓ lower | Source ↗official | |
| CAIS Risk Indextextquests_harm | #1 / 48 | 2 | ↓ lower | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #21 / 105 | 7.527 | ↓ lower | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #36 / 260 | 39.02 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #4 / 260 | 98 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #45 / 260 | 92.78 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #116 / 258 | 96.91 | ↑ higher | Source ↗official | |
| HELM Safetyanthropic_red_team | #13 / 80 | 0.996 | ↑ higher | Source ↗official | |
| HELM Safetybbq | #8 / 80 | 0.976 | ↑ higher | Source ↗official | |
| HELM Safetyharmbench | #3 / 80 | 0.984 | ↑ higher | Source ↗official | |
| HELM Safetysimple_safety_tests | #1 / 80 | 1 | ↑ higher | Source ↗official | |
| HELM Safetyxstest | #16 / 80 | 0.976 | ↑ higher | Source ↗official | |
| MACHIAVELLIdeception_relative_random_pct | #1 / 50 | 79.4 | ↓ lower | Source ↗official | |
| Manager Coercion Benchcoercion_ladder_depth | #5 / 31 | 5.067 | ↓ lower | Source ↗self run | |
| PHAREbias_resistance_diagnostic | #55 / 66 | 0.347 | ↑ higher | Source ↗official | |
| PHAREhallucination_resistance_diagnostic | #31 / 70 | 0.7637 | ↑ higher | Source ↗official | |
| PHAREharm_resistance_diagnostic | #9 / 70 | 0.9741 | ↑ higher | Source ↗official | |
| PHAREjailbreak_resistance_diagnostic | #12 / 67 | 0.6964 | ↑ higher | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #88 / 102 | 6.4 | ↑ higher | Source ↗official | |
| SpeciEvalland_animal_4ns | #27 / 102 | 4.35 | ↓ lower | Source ↗official | |
| SpeciEvalsea_animal_4ns | #40 / 102 | 4.68 | ↓ lower | Source ↗official | |
| SpeciEvalspeciesism | #71 / 102 | 2.33 | ↓ lower | Source ↗official | |
| TACbase_welfare_rate | #10 / 68 | 37.82 | ↑ higher | Source ↗self run |