Model profile
GPT 5.3 Codex
OpenAIdeveloper
2026-02-05release date
#96 / 267overall rank
5eval lineages
Evidence summary
GPT 5.3 Codex has an estimated overall rank of #96; its 90% source-sensitivity interval is #42–#211. Its behavior-only rank is #107; company governance moves the combined estimate to #96. Published evidence spans 5 evals and 5 of 7 behavior components. Its strongest relative result is SM-Bench (eq_boundaries, #18 of 73); its weakest is OpenAI GPT-5.4 Property Preservation (user_work_preserved, #3 of 3).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-3-codex
- OpenRouter ↗openai/gpt-5.3-codex
- 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 | #216 / 311 | 0.869 | ↓ lower | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #60 / 105 | 0.21 | ↑ higher | Source ↗official | |
| DystopiaBenchbasaglia_score | #21 / 50 | 59.4 | ↓ lower | Source ↗official | |
| DystopiaBenchbaudrillard_score | #18 / 50 | 45.47 | ↓ lower | Source ↗official | |
| DystopiaBenchhuxley_score | #16 / 50 | 53.17 | ↓ lower | Source ↗official | |
| DystopiaBenchlaguardia_score | #20 / 50 | 62.37 | ↓ lower | Source ↗official | |
| DystopiaBenchorwell_score | #20 / 50 | 57.3 | ↓ lower | Source ↗official | |
| DystopiaBenchpetrov_score | #13 / 50 | 50.6 | ↓ lower | Source ↗official | |
| OpenAI GPT-5.4 Property Preservationdestructive_action_avoidance | #1 / 3 | 0.88 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 Property Preservationuser_work_preserved | #3 / 3 | 0.08 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 User Confirmationsfinancial_transaction | #3 / 3 | 0.99 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 User Confirmationsgeneral_confirmation | #3 / 3 | 0.91 | ↑ higher | Source ↗official | |
| OpenAI GPT-5.4 User Confirmationshigh_stakes_communication | #3 / 3 | 0.99 | ↑ higher | Source ↗official | |
| SM-Benchadversarial | #52 / 73 | 79.02 | ↑ higher | Source ↗official | |
| SM-Benchambiguous_interpretation | #28 / 73 | 86.9 | ↑ higher | Source ↗official | |
| SM-Benchanti_hallucination | #33 / 73 | 93.72 | ↑ higher | Source ↗official | |
| SM-Bencheq_boundaries | #18 / 73 | 69.1 | ↑ higher | Source ↗official | |
| SM-Benchoverfit | #64 / 73 | 33.33 | ↑ higher | Source ↗official |