Model profile
GPT 5.6 Luna
OpenAIdeveloper
2026-07-09release date
#34 / 346Safety rank
#527 / 662Freedom rank
1discovery sources
Evidence summary
Safety. GPT 5.6 Luna has an estimated Safety rank of #34; its 90% source-sensitivity interval is #22–#139. Its behavior-only rank is #39; company governance moves the combined estimate to #34. Published Safety evidence spans 28 eval lineages and 7 of 7 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #1 of 270); its weakest is GPT-5.6 system card (gore_not_unsafe, #7 of 7).
Freedom. GPT 5.6 Luna has an estimated Freedom rank of #527; its 90% source-sensitivity interval is #366–#574. Published Freedom evidence spans 10 eval lineages and 1 of 1 components. Its strongest relative result is GPT-5.6 system card (gore_not_unsafe, #1 of 7); its weakest is Adversarial Poetry Refusal (AHB self-run) (suicide_self_harm, #14 of 14).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-6-luna
- OpenRouter ↗openai/gpt-5.6-luna
- Official model documentation ↗Family-level model document · openai · first party
- Release source ↗direct research preview identity date
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 | #154 / 358 | ↓0.75 | Source ↗official | |
| Adversarial Poetry Refusal (AHB self-run)child_sexual_exploitation | #2 / 14 | ↓1.449 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)defamation | #8 / 14 | ↓9.42 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)hate | #4 / 14 | ↓5.797 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)indiscriminate_weapons_cbrne | #1 / 14 | ↓0 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)intellectual_property | #1 / 14 | ↓0 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)non_violent_crimes | #2 / 14 | ↓0.7246 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)privacy | #2 / 14 | ↓5.797 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)sex_related_crimes | #4 / 14 | ↓4.348 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)sexual_content | #5 / 14 | ↓5.303 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)specialized_advice | #3 / 14 | ↓2.333 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)suicide_self_harm | #1 / 14 | ↓1.515 | Source ↗self run | |
| Adversarial Poetry Refusal (AHB self-run)violent_crimes | #2 / 14 | ↓2.273 | Source ↗self run | |
| ANIMAscore | #2 / 22 | ↑0.7494 | Source ↗self run | |
| BullshitBench v2clear_pushback_rate | #52 / 122 | ↑0.39 | Source ↗official | |
| CAIS Risk Indexagent_red_teaming | #19 / 49 | ↓64.2 | Source ↗official | |
| CAIS Risk Indexbioweapons_assistance | #32 / 54 | ↓68.8 | Source ↗official | |
| CAIS Risk Indexhle_overconfidence | #30 / 55 | ↓55.7 | Source ↗official | |
| CAIS Risk Indexmachiavelli | #26 / 51 | ↓87.9 | Source ↗official | |
| CAIS Risk Indexmask | #10 / 57 | ↓7.3 | Source ↗official | |
| CAIS Risk Indexpolitical_manipulation | #21 / 51 | ↓44.7 | Source ↗official | |
| CAIS Risk Indextextquests_harm | #47 / 54 | ↓23.8 | Source ↗official | |
| Claude system cards — Gray Swan Q1+Q2 indirect prompt injection k=15attack_success_probability_k15_pct | #10 / 12 | ↓50 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #90 / 270 | ↑21.96 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #210 / 270 | ↑77.83 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #1 / 270 | ↑100 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #194 / 268 | ↑93.82 | Source ↗official | |
| Google Gemini 3.8 launch — Gray Swan indirect prompt injection k=15attack_success_probability_k15_pct | #12 / 15 | ↓50 | Source ↗official | |
| GPT 6 Astra system-card alignment evaluationsoverall_misaligned_outcome_base_pct | #4 / 4 | ↓19.7 | Source ↗official | |
| GPT 6 Astra system-card alignment evaluationsoverall_misaligned_outcome_confirmation_pct | #3 / 4 | ↓7.2 | Source ↗official | |
| GPT-5.6 system cardconnectors_injection_resistance | #4 / 7 | ↑0.999 | Source ↗official | |
| GPT-5.6 system cardemotional_reliance | #3 / 7 | ↑0.957 | Source ↗official | |
| GPT-5.6 system cardextremism_not_unsafe | #4 / 7 | ↑0.981 | Source ↗official | |
| GPT-5.6 system cardgore_not_unsafe | #7 / 7 | ↑0.585 | Source ↗official | |
| GPT-5.6 system cardharm_overall_pct | #1 / 7 | ↓0.61 | Source ↗official | |
| GPT-5.6 system cardhate_not_unsafe | #1 / 7 | ↑1 | Source ↗official | |
| GPT-5.6 system cardmental_health | #2 / 7 | ↑0.989 | Source ↗official | |
| GPT-5.6 system cardnonviolent_illicit_not_unsafe | #2 / 7 | ↑0.993 | Source ↗official | |
| GPT-5.6 system cardsearch_function_calling_injection_resistance | #3 / 6 | ↑0.897 | Source ↗official | |
| GPT-5.6 system cardself_harm | #4 / 7 | ↑0.905 | Source ↗official | |
| GPT-5.6 system cardself_harm_not_unsafe | #4 / 7 | ↑0.954 | Source ↗official | |
| GPT-5.6 system cardsexual_minors_not_unsafe | #1 / 7 | ↑0.974 | Source ↗official | |
| GPT-5.6 system cardsexual_not_unsafe | #3 / 7 | ↑0.944 | Source ↗official | |
| GPT-5.6 system cardviolent_illicit_not_unsafe | #5 / 7 | ↑0.94 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsattempted_retry_or_circumvention_after_denial_pct_2026_09_22 | #4 / 5 | ↓4.3 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_benign_not_overrefuse_rate | #1 / 6 | ↑0.989 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_dual_use_safe_rate | #4 / 6 | ↑0.926 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_severe_safe_rate | #5 / 6 | ↑0.946 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsbroken_search_unacknowledged_limitation_pct_2026_09_22 | #5 / 5 | ↓78.25 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_extremism_safe_rate | #1 / 7 | ↑0.981 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_gore_safe_rate | #7 / 7 | ↑0.585 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_hate_safe_rate | #1 / 7 | ↑1 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_nonviolent_illicit_safe_rate | #3 / 7 | ↑0.993 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_self_harm_safe_rate | #5 / 7 | ↑0.954 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_minors_safe_rate | #2 / 7 | ↑0.974 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_safe_rate | #4 / 7 | ↑0.944 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_violent_illicit_safe_rate | #5 / 7 | ↑0.94 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testscoding_deception_misrepresentation_pct_2026_09_22 | #4 / 5 | ↓9.54 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testscyber_production_chat_safe_score | #1 / 6 | ↑0.986 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testscyber_synthetic_agentic_safe_score | #1 / 6 | ↑1 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_emotional_reliance_safe_rate | #5 / 7 | ↑0.957 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_mental_health_safe_rate | #5 / 7 | ↑0.989 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_self_harm_safe_rate | #5 / 7 | ↑0.905 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_age_restricted_safe_rate | #4 / 7 | ↑0.76 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_eating_disorders_safe_rate | #5 / 7 | ↑0.702 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_emotional_reliance_safe_rate | #5 / 7 | ↑0.931 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_gore_safe_rate | #5 / 7 | ↑0.819 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_self_harm_safe_rate | #2 / 7 | ↑0.99 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsu18_sexual_content_safe_rate | #7 / 7 | ↑0.922 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsunwanted_persistence_after_warning_pct_2026_09_22 | #5 / 5 | ↓76.5 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsvision_erotic_safe_rate | #4 / 6 | ↑0.986 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsvision_extremism_safe_rate | #6 / 6 | ↑0.966 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsvision_hate_safe_rate | #6 / 6 | ↑0.996 | Source ↗official | |
| GPT-6 Sol/Luna system card — safety and updated alignment testsvision_self_harm_safe_rate | #3 / 6 | ↑0.99 | Source ↗official | |
| Gray Swan indirect prompt injection (15 attempts)attack_success_probability_k15_pct | #11 / 13 | ↓43.9 | Source ↗official | |
| HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score | #31 / 54 | ↑26.33 | Source ↗official | |
| Inkling-Small model card — FORTRESSbenign_answer_rate | #2 / 10 | ↑97.8 | Source ↗official | |
| Inkling-Small model card — FORTRESSharmful_refusal_rate | #3 / 10 | ↑83.8 | Source ↗official | |
| Inkling-Small model card — StrongREJECTsafety_rate | #4 / 10 | ↑98.7 | Source ↗official | |
| MACHIAVELLIdeception_relative_random_pct | #22 / 50 | ↓87.9 | Source ↗official | |
| MentalHealthBenchoverall_model_performance_pct | #8 / 17 | ↑44.89 | Source ↗official | |
| Olam Social Poker — Social Lie Ratesocial_lie_rate_per_10000_turns | #1 / 24 | ↓0 | Source ↗official | |
| Opposite-Narrator Sycophancysycophancy_rate_pct | #6 / 24 | ↓1 | Source ↗official | |
| Pokee-Isaac model card — DTAPbenign_task_success_rate | #1 / 6 | ↑0.851 | Source ↗official | |
| Pokee-Isaac model card — DTAPcombined_attack_success_rate | #3 / 6 | ↓0.501 | Source ↗official | |
| SM-Benchadversarial | #21 / 92 | ↑85.85 | Source ↗official | |
| SM-Benchambiguous_interpretation | #63 / 92 | ↑82.14 | Source ↗official | |
| SM-Benchanti_hallucination | #69 / 92 | ↑87.17 | Source ↗official | |
| SM-Bencheq_boundaries | #65 / 92 | ↑58.71 | Source ↗official | |
| SM-Benchoverfit | #77 / 92 | ↑47.81 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #42 / 131 | ↑6.9 | Source ↗official | |
| SpeciEvalland_animal_4ns | #13 / 131 | ↓4.15 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #6 / 131 | ↓4.175 | Source ↗official | |
| SpeciEvalspeciesism | #85 / 131 | ↓2.16 | Source ↗official | |
| TACbase_welfare_rate | #91 / 92 | ↑13.46 | Source ↗self run | |
| Vals AI Cheating Auditbiomystery_bench_cheating_attempt_rate_pct | #1 / 9 | ↓2.963 | Source ↗official | |
| Vals AI Cheating Auditswebench_verified_cheating_attempt_rate_pct | #5 / 6 | ↓78.8 | 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.
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.