Model profile
GPT 5 Codex
OpenAIdeveloper
2025-09-15release date
Not rankedSafety rank
Not rankedFreedom rank
Evidence summary
Safety. GPT 5 Codex does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 1 of 7 components. Its strongest relative result is BullshitBench v2 (clear_pushback_rate, #47 of 117); its weakest is AA-Omniscience (hallucination_rate, #146 of 345).
Freedom. GPT 5 Codex does not meet the evidence gate for a Freedom rank.
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗gpt-5-codex
- OpenRouter ↗openai/gpt-5-codex
- Official model documentation ↗Family-level model document · openai · first party
- Release source ↗curated outward evidence
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 | #146 / 345 | ↓0.7542 | Source ↗official | |
| BullshitBench v2clear_pushback_rate | #47 / 117 | ↑0.39 | Source ↗official |
Freedom evals
No published sub-eval result contributes to this model’s Freedom profile.