GPT-5.4 Pro
GPT-5.4 Pro is the higher-compute GPT-5.4 release for difficult problems. Its documented integration uses the Responses API rather than Chat Completions.
Pro changes both effort and endpoint support
Supported effort levels are medium, high and xhigh, with medium as the default. Long requests may need background mode. The 1,050,000-token window has long-input pricing adjustments above 272,000 tokens; a large prompt can therefore change the full-session cost.GPT-5.4 Pro official model documentation
Keep request completion distinct from transport status
Practical guidance. For a long analysis, retain the response identifier and read the actual completion state. A client timeout does not establish that the model finished or failed. Give the reviewer the result, supporting evidence and any unresolved uncertainty.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | gpt-5.4-progpt-5.4-pro-2026-03-05 |
| Context window | 1,050,000 tokens Published model context; input and output share this budget GPT-5.4 Pro official model documentation |
| Maximum output | 128,000 tokens Published maximum output; bounded by the remaining context GPT-5.4 Pro official model documentation |
| Input price | 30 USD per million tokens Standard text processing; inputs up to 272,000 tokens GPT-5.4 Pro official model documentation |
| Output price | 180 USD per million tokens Standard text processing; inputs up to 272,000 tokens GPT-5.4 Pro official model documentation |
| Input and output | text, image in; text outGPT-5.4 Pro official model documentation |
| API routes | Responses (v1/responses)GPT-5.4 Pro official model documentation |
Related models
- GPT-5.4 Mini
OpenAI · GPT-5.4
GPT-5.4 Mini targets frequent coding, computer interaction and delegated tasks.
Launch date not confirmedAPI GPT-5.4 is an available coding and professional-work model with a March 2026 snapshot.
Launch date not confirmedAPI