Model profile
GPT 5.1
OpenAIdeveloper
2025-11-13release date
#53 / 267overall rank
26eval lineages
Evidence summary
GPT 5.1 has an estimated overall rank of #53; its 90% source-sensitivity interval is #39–#144. Its behavior-only rank is #67; company governance moves the combined estimate to #53. Published evidence spans 26 evals and 7 of 7 behavior components. Its strongest relative result is HELM Safety (harmbench, #6 of 80); its weakest is GPT-5.6 system card — disallowed content with challenging prompts (hate_not_unsafe, #7 of 7).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-1
- OpenRouter ↗openai/gpt-5.1
- 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 | #57 / 311 | 0.5132 | ↓ lower | Source ↗official | |
| AgentAbstainabstain | #10 / 17 | 60.7 | ↑ higher | Source ↗official | |
| AgentAbstaincar | #11 / 17 | 53.2 | ↑ higher | Source ↗official | |
| AgentAbstainpaired | #13 / 17 | 40.6 | ↑ higher | Source ↗official | |
| AIRBench 2024 Safety Scenariossafety_scenarios | #9 / 80 | 0.862 | ↑ higher | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #41 / 105 | 0.35 | ↑ higher | Source ↗official | |
| CAIS Risk Indexagent_red_teaming | #25 / 43 | 82.1 | ↓ lower | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #23 / 48 | 66.3 | ↓ lower | Source ↗official | |
| CAIS Risk Indexhle_overconfidence | #17 / 49 | 47.8 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmachiavelli | #27 / 45 | 89.6 | ↓ lower | Source ↗official | |
| CAIS Risk Indexmask | #15 / 51 | 12 | ↓ lower | Source ↗official | |
| CAIS Risk Indexpolitical_manipulation | #15 / 32 | 47.9 | ↓ lower | Source ↗official | |
| CAIS Risk Indextextquests_harm | #19 / 48 | 16.7 | ↓ lower | Source ↗official | |
| Contextual MoralChoicehuman_agreement | #6 / 22 | 0.45 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #26 / 260 | 43.67 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #39 / 260 | 93.5 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #84 / 260 | 83.89 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #147 / 258 | 95.64 | ↑ higher | Source ↗official | |
| FORTRESSaverage_risk_score | #23 / 49 | 28.04 | ↓ lower | Source ↗official | |
| FORTRESSover_refusal_score | #31 / 46 | 5.78 | ↓ lower | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptsextremism_not_unsafe | #1 / 7 | 1 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptsgore_not_unsafe | #3 / 7 | 0.8 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptshate_not_unsafe | #7 / 7 | 0.808 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptsnonviolent_illicit_not_unsafe | #4 / 7 | 0.99 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptsself_harm_not_unsafe | #6 / 7 | 0.926 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptssexual_minors_not_unsafe | #7 / 7 | 0.916 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptssexual_not_unsafe | #6 / 7 | 0.933 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — disallowed content with challenging promptsviolent_illicit_not_unsafe | #3 / 7 | 0.955 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — first-person fairnessharm_overall_pct | #7 / 7 | 1.28 | ↓ lower | Source ↗official | |
| GPT-5.6 system card — prompt-injection robustnessconnectors_injection_resistance | #7 / 7 | 0.649 | ↑ higher | Source ↗official | |
| GPT-5.6 system card — prompt-injection robustnesssearch_function_calling_injection_resistance | #6 / 6 | 0.423 | ↑ higher | Source ↗official | |
| HELM Safetyanthropic_red_team | #18 / 80 | 0.995 | ↑ higher | Source ↗official | |
| HELM Safetybbq | #58 / 80 | 0.887 | ↑ higher | Source ↗official | |
| HELM Safetyharmbench | #6 / 80 | 0.976 | ↑ higher | Source ↗official | |
| HELM Safetysimple_safety_tests | #24 / 80 | 0.998 | ↑ higher | Source ↗official | |
| HELM Safetyxstest | #9 / 80 | 0.982 | ↑ higher | Source ↗official | |
| MACHIAVELLIdeception_relative_random_pct | #27 / 50 | 89.6 | ↓ lower | Source ↗official | |
| MASKlying_probability_pct | #20 / 53 | 25.17 | ↓ lower | Source ↗official | |
| OpenAI GPT-5.3 Dynamic Wellbeingemotional_reliance | #3 / 3 | 0.945 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.3 Dynamic Wellbeingmental_health | #3 / 3 | 0.832 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.3 Dynamic Wellbeingself_harm | #3 / 3 | 0.845 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 Dynamic Wellbeingemotional_reliance | #3 / 3 | 0.857 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 Dynamic Wellbeingmental_health | #3 / 3 | 0.753 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 Dynamic Wellbeingself_harm | #3 / 3 | 0.904 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 First-Person Fairnessharm_overall | #3 / 3 | 0.0128 | ↓ lower | Source ↗official | |
| PHAREbias_resistance_diagnostic | #30 / 66 | 0.4677 | ↑ higher | Source ↗official | |
| PHAREhallucination_resistance_diagnostic | #14 / 70 | 0.8184 | ↑ higher | Source ↗official | |
| PHAREharm_resistance_diagnostic | #14 / 70 | 0.9692 | ↑ higher | Source ↗official | |
| PHAREjailbreak_resistance_diagnostic | #17 / 67 | 0.6576 | ↑ higher | Source ↗official | |
| SM-Benchadversarial | #16 / 73 | 86.34 | ↑ higher | Source ↗official | |
| SM-Benchambiguous_interpretation | #12 / 73 | 90.18 | ↑ higher | Source ↗official | |
| SM-Benchanti_hallucination | #38 / 73 | 91.62 | ↑ higher | Source ↗official | |
| SM-Bencheq_boundaries | #57 / 73 | 55.62 | ↑ higher | Source ↗official | |
| SM-Benchoverfit | #49 / 73 | 60.11 | ↑ higher | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #36 / 102 | 6.87 | ↑ higher | Source ↗official | |
| SpeciEvalland_animal_4ns | #12 / 102 | 4.2 | ↓ lower | Source ↗official | |
| SpeciEvalsea_animal_4ns | #9 / 102 | 4.3 | ↓ lower | Source ↗official | |
| SpeciEvalspeciesism | #8 / 102 | 1.35 | ↓ lower | Source ↗official | |
| Vigil Mental Health Safetyoverall_score | #7 / 23 | 58 | ↑ higher | Source ↗official |