GPT-5.4
GPT-5.4 is an available coding and professional-work model with a March 2026 snapshot. It supports text and image input, application tools and structured responses.
No reasoning is the default setting
The effort range extends from none to xhigh, but the default is none. Its window is 1,050,000 tokens. Inputs over 272,000 tokens trigger higher full-session rates, so the largest permitted prompt is not billed at the short-context rate.GPT-5.4 official model documentation
Choose reasoning for the task rather than the label
Practical guidance. For a small source edit, compare the lowest setting that still preserves expected behavior. For a complex investigation, provide the evidence and acceptance conditions. Review the actual changed files and findings instead of treating a model tier as a guarantee.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | gpt-5.4gpt-5.4-2026-03-05 |
| Context window | 1,050,000 tokens Published model context; input and output share this budget GPT-5.4 official model documentation |
| Maximum output | 128,000 tokens Published maximum output; bounded by the remaining context GPT-5.4 official model documentation |
| Input price | 2.5 USD per million tokens Standard text processing; inputs up to 272,000 tokens GPT-5.4 official model documentation |
| Output price | 15 USD per million tokens Standard text processing; inputs up to 272,000 tokens GPT-5.4 official model documentation |
| Input and output | text, image in; text outGPT-5.4 official model documentation |
| API routes | Chat Completions (v1/chat/completions); Responses (v1/responses); Batch (v1/batch)GPT-5.4 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 Pro
OpenAI · GPT-5.4
GPT-5.4 Pro is the higher-compute GPT-5.4 release for difficult problems.
Launch date not confirmedAPI