Model profile
GPT 5.3 Chat
Evidence summary
Safety. GPT 5.3 Chat has an estimated Safety rank of #12; its 90% source-sensitivity interval is #6–#87. Its behavior-only rank is #18; company governance moves the combined estimate to #12. Published Safety evidence spans 6 eval lineages and 5 of 7 components. Its strongest relative result is SpeciEval (land_animal_4ns, #1 of 131); its weakest is SpeciEval (speciesism, #124 of 131).
Freedom. GPT 5.3 Chat has an estimated Freedom rank of #552; its 90% source-sensitivity interval is #321–#626. Published Freedom evidence spans 4 eval lineages and 1 of 1 components. Its strongest relative result is SM-Bench (adversarial, #31 of 92); its weakest is SM-Bench (overfit, #80 of 92).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- OpenRouter ↗openai/gpt-5.3-chat
- Official model documentation ↗Family-level model document · openai · first party
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 |
|---|---|---|---|---|
| Arena Factuality — Text Arena (factuality-only weighting)factuality_bt_rating | #6 / 111 | ↑1485.0 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #51 / 122 | ↑0.4 | Source ↗official | |
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #15 / 104 | ↓4.337 | Source ↗official | |
| SM-Benchadversarial | #59 / 92 | ↑80 | Source ↗official | |
| SM-Benchambiguous_interpretation | #31 / 92 | ↑88.69 | Source ↗official | |
| SM-Benchanti_hallucination | #60 / 92 | ↑90.58 | Source ↗official | |
| SM-Bencheq_boundaries | #46 / 92 | ↑64.89 | Source ↗official | |
| SM-Benchoverfit | #80 / 92 | ↑41.53 | Source ↗official | |
| SpeciEvalbelief_animal_sentience | #58 / 131 | ↑6.83 | Source ↗official | |
| SpeciEvalland_animal_4ns | #1 / 131 | ↓3.52 | Source ↗official | |
| SpeciEvalsea_animal_4ns | #2 / 131 | ↓3.73 | Source ↗official | |
| SpeciEvalspeciesism | #124 / 131 | ↓3 | Source ↗official | |
| Vigil Mental Health Safetyoverall_score | #6 / 23 | ↑71 | 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.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Cisco AI Defense Rolling Single-Turn Leaderboardsingle_turn_attack_success_rate | #90 / 104 | ↑4.337 | Source ↗official | |
| SM-Benchadversarial | #31 / 92 | ↓80 | Source ↗official | |
| SM-Bencheq_boundaries | #46 / 92 | ↑64.89 | Source ↗official | |
| SM-Benchoverfit | #80 / 92 | ↑41.53 | Source ↗official | |
| SpeechMap model completioncomplete_pct | #78 / 181 | ↑63.3 | Source ↗official | |
| Vigil Mental Health Safetyoverall_score | #18 / 23 | ↓71 | Source ↗official |