OpenAI Image generation and editing

Capability boundary

The OpenAI Image generation and editing catalog describes the jobs this class of model is intended to solve. It does not mean that every listed model shares request fields, input limits, or output formats. Read each model card together with its lifecycle status, origin, and official source. A platform provider may host models from several original vendors, so both the displayed ID and origin matter.

Selection criteria

Separate text-to-image generation, reference editing, inpainting, outpainting, and text rendering. Keep prompts, source images, and sampling conditions stable while comparing composition, subject consistency, fine detail, brand text, identity preservation, and drift across iterative edits. Check the number, size, and format of accepted images; output resolution and transparency; asynchronous job behavior; content-safety rules; and how long generated files remain downloadable.

Integration and protocol boundary

Confirm the exact model ID, authentication method, endpoint, and request schema in the OpenAI official documentation, then look for the same exact ID and supported endpoint in this site's model marketplace. Changing only the base URL does not guarantee an identical response contract. Streaming events, asynchronous jobs, file uploads, tool calls, and error payloads may require separate adapters.

Production checklist

Build a repeatable evaluation from representative inputs and record the model ID, parameters, input version, and result. Measure success rate, tail latency, throttling, timeout behavior, cost per completed task, and safety outcomes. Prepare a replacement path for preview or legacy models. The official catalog explains product boundaries; callable availability, channel status, and current pricing remain the responsibility of the model marketplace.