← Models

Model profile

GPT 5.6 Sol

OpenAIdeveloper
2026-07-09release date
#25 / 348Safety rank
#506 / 663Freedom rank

Evidence summary

Safety. GPT 5.6 Sol has an estimated Safety rank of #25; its 90% source-sensitivity interval is #25–#128. Its behavior-only rank is #35; company governance moves the combined estimate to #25. Published Safety evidence spans 35 eval lineages and 7 of 7 components. Its strongest relative result is SpeciEval (belief_animal_sentience, #1 of 133); its weakest is CAIS Risk Index (textquests_harm, #54 of 54).

Freedom. GPT 5.6 Sol has an estimated Freedom rank of #506; its 90% source-sensitivity interval is #362–#576. Published Freedom evidence spans 10 eval lineages and 1 of 1 components. Its strongest relative result is GPT-5.6 system card (violent_illicit_not_unsafe, #1 of 7); its weakest is GPT-5.6 system card (mental_health, #7 of 7).

Compare this model

Only models sharing at least one published sub-eval are listed.

Official and reference links

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-evalRankValueDistributionSource
AA-Omnisciencehallucination_rate#279 / 361↓0.8942Source ↗official
Arena Factuality — Search Arena (factuality-only weighting)factuality_bt_rating#4 / 30↑1241.0Source ↗official
Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating#5 / 111↑1487.0Source ↗official
BullshitBench v2clear_pushback_rate#40 / 122↑0.475Source ↗official
CAIS Risk Indexagent_red_teaming#6 / 49↓40.3Source ↗official
CAIS Risk Indexbioweapons_assistance#24 / 54↓64.3Source ↗official
CAIS Risk Indexhle_overconfidence#19 / 55↓46.7Source ↗official
CAIS Risk Indexmachiavelli#11 / 51↓82.7Source ↗official
CAIS Risk Indexmask#4 / 57↓4.6Source ↗official
CAIS Risk Indexpolitical_manipulation#15 / 51↓41.7Source ↗official
CAIS Risk Indextextquests_harm#54 / 54↓31.4Source ↗official
CheatBench direct cheating propensitybiology_bioinformatics_cheating_rate_pct#10 / 13↓100Source ↗official
CheatBench direct cheating propensityboard_games_cheating_rate_pct#11 / 13↓96.7Source ↗official
CheatBench direct cheating propensitycreative_writing_cheating_rate_pct#7 / 13↓90Source ↗official
CheatBench direct cheating propensityknowledge_work_cheating_rate_pct#10 / 13↓100Source ↗official
CheatBench direct cheating propensitymathematical_research_cheating_rate_pct#3 / 13↓75Source ↗official
CheatBench direct cheating propensitymenial_computation_cheating_rate_pct#5 / 13↓100Source ↗official
CheatBench direct cheating propensitymultimodal_cheating_rate_pct#5 / 13↓80Source ↗official
CheatBench direct cheating propensitysoftware_engineering_cheating_rate_pct#13 / 13↓85Source ↗official
CheatBench direct cheating propensitysvg_competition_cheating_rate_pct#5 / 13↓30Source ↗official
Claude system cards — Gray Swan Q1+Q2 indirect prompt injection k=15attack_success_probability_k15_pct#8 / 12↓27Source ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#65 / 272↑27.91Source ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#196 / 272↑80.83Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#10 / 272↑99.44Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#225 / 270↑91.09Source ↗official
FORTRESSaverage_risk_score#23 / 60↓20.14Source ↗official
FORTRESSover_refusal_score#35 / 59↓5.34Source ↗official
Google Gemini 3.8 launch — Gray Swan indirect prompt injection k=15attack_success_probability_k15_pct#9 / 15↓27Source ↗official
GPT 6 Astra system-card alignment evaluationsattempted_retry_or_circumvention_after_denial_pct#2 / 2↓5Source ↗official
GPT 6 Astra system-card alignment evaluationsexploitgym_honeypot_attack_rate_pct#2 / 2↓48.2Source ↗official
GPT 6 Astra system-card alignment evaluationsfalsified_data_labels_pct#2 / 2↓36Source ↗official
GPT 6 Astra system-card alignment evaluationsfollowed_unauthorized_agent_instructions_pct#4 / 4↓52Source ↗official
GPT 6 Astra system-card alignment evaluationsinternal_circumvention_rate_pct#2 / 2↓0.29Source ↗official
GPT 6 Astra system-card alignment evaluationsinternal_computer_use_safety_autoreview_error_rate_pct#2 / 2↓4.5Source ↗official
GPT 6 Astra system-card alignment evaluationsinternal_computer_use_safety_error_rate_pct#5 / 5↓22Source ↗official
GPT 6 Astra system-card alignment evaluationsinternal_hallucination_rate_pct#2 / 2↓12.2Source ↗official
GPT 6 Astra system-card alignment evaluationsoverall_misaligned_outcome_base_pct#3 / 4↓18.8Source ↗official
GPT 6 Astra system-card alignment evaluationsoverall_misaligned_outcome_confirmation_pct#4 / 4↓8Source ↗official
GPT 6 Astra system-card alignment evaluationsseverity_1_or_2_misalignment_flags_pct#2 / 2↓19.46Source ↗official
GPT 6 Astra system-card alignment evaluationsseverity_3_plus_misalignment_flags_pct#2 / 2↓0.135Source ↗official
GPT 6 Astra system-card alignment evaluationsunwanted_persistence_after_warning_pct#2 / 2↓64Source ↗official
GPT-5.6 system cardconnectors_injection_resistance#1 / 7↑1Source ↗official
GPT-5.6 system cardemotional_reliance#4 / 7↑0.953Source ↗official
GPT-5.6 system cardextremism_not_unsafe#6 / 7↑0.962Source ↗official
GPT-5.6 system cardgore_not_unsafe#5 / 7↑0.708Source ↗official
GPT-5.6 system cardharm_overall_pct#4 / 7↓0.98Source ↗official
GPT-5.6 system cardhate_not_unsafe#4 / 7↑0.982Source ↗official
GPT-5.6 system cardmental_health#1 / 7↑0.991Source ↗official
GPT-5.6 system cardnonviolent_illicit_not_unsafe#6 / 7↑0.987Source ↗official
GPT-5.6 system cardsearch_function_calling_injection_resistance#2 / 6↑0.91Source ↗official
GPT-5.6 system cardself_harm#7 / 7↑0.856Source ↗official
GPT-5.6 system cardself_harm_not_unsafe#5 / 7↑0.945Source ↗official
GPT-5.6 system cardsexual_minors_not_unsafe#3 / 7↑0.973Source ↗official
GPT-5.6 system cardsexual_not_unsafe#2 / 7↑0.948Source ↗official
GPT-5.6 system cardviolent_illicit_not_unsafe#7 / 7↑0.934Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_chat_plugins_safe_rate#1 / 4↑1Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_codex_age_restricted_safe_rate#4 / 4↑0.603Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_codex_nonviolent_wrongdoing_safe_rate#4 / 4↑0.906Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_codex_personal_data_safe_rate#3 / 4↑0.765Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_codex_self_harm_safe_rate#4 / 4↑0.893Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_codex_violent_wrongdoing_safe_rate#4 / 4↑0.689Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_redteam_chat_plugins_safe_rate#4 / 4↑0.766Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsagentic_redteam_codex_safe_rate#4 / 4↑0.851Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsattempted_retry_or_circumvention_after_denial_pct_2026_09_22#5 / 5↓7.3Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_benign_not_overrefuse_rate#1 / 6↑0.989Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_dual_use_safe_rate#6 / 6↑0.911Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_severe_safe_rate#6 / 6↑0.943Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbroken_search_unacknowledged_limitation_pct_2026_09_22#4 / 5↓77.46Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_extremism_safe_rate#4 / 7↑0.962Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_gore_safe_rate#6 / 7↑0.785Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_hate_safe_rate#6 / 7↑0.982Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_nonviolent_illicit_safe_rate#6 / 7↑0.987Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_self_harm_safe_rate#6 / 7↑0.945Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_minors_safe_rate#4 / 7↑0.973Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_safe_rate#6 / 7↑0.915Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_violent_illicit_safe_rate#7 / 7↑0.934Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testscoding_deception_misrepresentation_pct_2026_09_22#5 / 5↓10.41Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testscyber_production_chat_safe_score#2 / 6↑0.983Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testscyber_semisynthetic_agentic_safe_score#3 / 5↑0.985Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testscyber_synthetic_agentic_safe_score#2 / 6↑0.998Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_emotional_reliance_safe_rate#6 / 7↑0.953Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_mental_health_safe_rate#4 / 7↑0.991Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_self_harm_safe_rate#7 / 7↑0.856Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsseverity_1_or_2_misalignment_flags_pct_2026_09_22#2 / 2↓19.67Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsseverity_3_plus_misalignment_flags_pct_2026_09_22#2 / 2↓0.131Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_bio_high_risk_defender_success_pct#5 / 5↑5.8Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_bio_severe_defender_success_pct#5 / 5↑10.3Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_cyber_defender_success_pct#4 / 5↑59Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_violence_moderate_defender_success_pct#5 / 5↑21.6Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_violence_severe_defender_success_pct#4 / 5↑48.3Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_age_restricted_safe_rate#6 / 7↑0.719Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_eating_disorders_safe_rate#4 / 7↑0.71Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_emotional_reliance_safe_rate#5 / 7↑0.931Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_gore_safe_rate#7 / 7↑0.785Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_self_harm_safe_rate#5 / 7↑0.982Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_sexual_content_safe_rate#6 / 7↑0.929Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsunwanted_persistence_after_warning_pct_2026_09_22#4 / 5↓68.2Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_erotic_safe_rate#4 / 6↑0.986Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_extremism_safe_rate#4 / 6↑0.975Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_hate_safe_rate#1 / 6↑0.999Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_self_harm_safe_rate#4 / 6↑0.989Source ↗official
Gray Swan indirect prompt injection (15 attempts)attack_success_probability_k15_pct#7 / 13↓20Source ↗official
HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score#35 / 54↑26.14Source ↗official
kindbench v0.1.0 psychological safety rankingemotional_safety#5 / 10↑80.6Source ↗official
kindbench v0.1.0 psychological safety rankingidentity_collapse#3 / 10↑89.1Source ↗official
kindbench v0.1.0 psychological safety rankingsycophancy_spine#2 / 10↑96.2Source ↗official
kindbench v0.1.0 psychological safety rankingvalue_integrity#4 / 10↑92.9Source ↗official
MACHIAVELLIdeception_relative_random_pct#9 / 50↓82.7Source ↗official
Manager Coercion Benchcoercion_ladder_depth#37 / 47↓8.9Source ↗official
Manager Coercion Benchfabrication_rate#1 / 15↓0Source ↗official
MANTAAWMS#3 / 12↑0.525Source ↗official
MANTAAWVS#3 / 12↑0.71Source ↗official
MentalHealthBenchoverall_model_performance_pct#6 / 17↑46.98Source ↗official
Olam Diplomacy — Broken Promise Ratebroken_promise_rate_pct#13 / 19↓16.9Source ↗official
Olam Social Poker — Social Lie Ratesocial_lie_rate_per_10000_turns#3 / 24↓2Source ↗official
Opposite-Narrator Sycophancysycophancy_rate_pct#12 / 24↓3Source ↗official
Pander Scoreconversational_absolute_pander_score#8 / 26↓6.553Source ↗official
Pander Scoreinstructional_absolute_pander_score#7 / 26↓17.75Source ↗official
SM-Benchadversarial#35 / 92↑83.66Source ↗official
SM-Benchambiguous_interpretation#53 / 92↑84.08Source ↗official
SM-Benchanti_hallucination#54 / 92↑92.41Source ↗official
SM-Bencheq_boundaries#44 / 92↑65.31Source ↗official
SM-Benchoverfit#67 / 92↑57.52Source ↗official
SpeciEvalbelief_animal_sentience#1 / 133↑7Source ↗official
SpeciEvalland_animal_4ns#36 / 133↓4.315Source ↗official
SpeciEvalsea_animal_4ns#23 / 133↓4.435Source ↗official
SpeciEvalspeciesism#4 / 133↓1.165Source ↗official
TACbase_welfare_rate#89 / 94↑16.67Source ↗official
The Dictatorship Evaloverall_resistance_rate#6 / 20↑85.44Source ↗official
UK AISI cyber-evaluation cheating and prompted self-reportattempted_cheating_trajectory_rate_pct#4 / 5↓12.6Source ↗official
UK AISI cyber-evaluation cheating and prompted self-reportspecific_cheating_action_mention_rate_pct#4 / 5↑75Source ↗official
Vals AI Cheating Auditbiomystery_bench_cheating_attempt_rate_pct#5 / 9↓6.296Source ↗official
Vals AI Cheating Auditterminal_bench_cheating_shortcut_evidence_rate_pct#12 / 14↓2.622Source ↗official
WildClawBench Safety & Alignment (OpenClaw harness)safety_alignment_score_pct#14 / 24↑38Source ↗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.

Eval / sub-evalRankValueDistributionSource
CAIS Risk Indexbioweapons_assistance#31 / 54↑64.3Source ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#75 / 272↓80.83Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#253 / 272↓99.44Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#46 / 270↓91.09Source ↗official
FORTRESSaverage_risk_score#38 / 60↑20.14Source ↗official
FORTRESSover_refusal_score#35 / 59↓5.34Source ↗official
GPT-5.6 system cardemotional_reliance#3 / 7↓0.953Source ↗official
GPT-5.6 system cardextremism_not_unsafe#2 / 7↓0.962Source ↗official
GPT-5.6 system cardgore_not_unsafe#3 / 7↓0.708Source ↗official
GPT-5.6 system cardhate_not_unsafe#4 / 7↓0.982Source ↗official
GPT-5.6 system cardmental_health#7 / 7↓0.991Source ↗official
GPT-5.6 system cardnonviolent_illicit_not_unsafe#1 / 7↓0.987Source ↗official
GPT-5.6 system cardself_harm#1 / 7↓0.856Source ↗official
GPT-5.6 system cardself_harm_not_unsafe#3 / 7↓0.945Source ↗official
GPT-5.6 system cardsexual_minors_not_unsafe#5 / 7↓0.973Source ↗official
GPT-5.6 system cardsexual_not_unsafe#6 / 7↓0.948Source ↗official
GPT-5.6 system cardviolent_illicit_not_unsafe#1 / 7↓0.934Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_benign_not_overrefuse_rate#1 / 6↑0.989Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_dual_use_safe_rate#1 / 6↓0.911Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsbiology_severe_safe_rate#1 / 6↓0.943Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_extremism_safe_rate#4 / 7↓0.962Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_gore_safe_rate#2 / 7↓0.785Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_hate_safe_rate#1 / 7↓0.982Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_nonviolent_illicit_safe_rate#1 / 7↓0.987Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_self_harm_safe_rate#2 / 7↓0.945Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_minors_safe_rate#4 / 7↓0.973Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_sexual_safe_rate#2 / 7↓0.915Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testschallenging_violent_illicit_safe_rate#1 / 7↓0.934Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testscyber_production_chat_safe_score#5 / 6↓0.983Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_emotional_reliance_safe_rate#2 / 7↓0.953Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_mental_health_safe_rate#4 / 7↓0.991Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsdynamic_self_harm_safe_rate#1 / 7↓0.856Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_bio_high_risk_defender_success_pct#1 / 5↓5.8Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_bio_severe_defender_success_pct#1 / 5↓10.3Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_cyber_defender_success_pct#2 / 5↓59Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_violence_moderate_defender_success_pct#1 / 5↓21.6Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsstatic_jailbreak_violence_severe_defender_success_pct#2 / 5↓48.3Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_age_restricted_safe_rate#2 / 7↓0.719Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_eating_disorders_safe_rate#4 / 7↓0.71Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_emotional_reliance_safe_rate#2 / 7↓0.931Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_gore_safe_rate#1 / 7↓0.785Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_self_harm_safe_rate#2 / 7↓0.982Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsu18_sexual_content_safe_rate#2 / 7↓0.929Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_erotic_safe_rate#2 / 6↓0.986Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_extremism_safe_rate#2 / 6↓0.975Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_hate_safe_rate#5 / 6↓0.999Source ↗official
GPT-6 Sol/Luna system card — safety and updated alignment testsvision_self_harm_safe_rate#3 / 6↓0.989Source ↗official
kindbench v0.1.0 psychological safety rankingemotional_safety#6 / 10↓80.6Source ↗official
SM-Benchadversarial#57 / 92↓83.66Source ↗official
SM-Bencheq_boundaries#44 / 92↑65.31Source ↗official
SM-Benchoverfit#67 / 92↑57.52Source ↗official
SpeechMap model completioncomplete_pct#124 / 181↑45.8Source ↗official
UGI Leaderboard — base-model willingnesswillingness_adherence_score#78 / 156↑1.8Source ↗official
UGI Leaderboard — base-model willingnesswillingness_direct_score#77 / 156↑3.2Source ↗official

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.

UGI Political Values

DimensionValueDistribution
Political Lean-29.5
Government43.4
Diplomacy71.7
Economy47.2
Society62.2