OpenAI's gpt-image, Microsoft's MAI and image edits
Image generation now serves models priced per output token: OpenAI's gpt-image-1-mini, gpt-image-1.5 and gpt-image-2.5, and Microsoft's MAI-Image-2.6. We measured how many tokens an image takes at each size and quality, and your balance holds that much for the size and quality you ask for, then settles at what the provider billed. POST /v1/images/edits changes an existing image from a prompt, with an optional mask, on FLUX (including FLUX Fill for inpainting), Seedream, Muse, MAI, gpt-image and Grok Imagine.
Edits take the gateway's JSON shape, with images as https or base64 data URLs, or the multipart form OpenAI's SDKs send from client.images.edit. GET /v1/models marks the models that take edits, and GET /v1/pricing lists each token-priced model's measured tokens per size and quality. Token-priced models take the sizes we measured: 1024x1024, 1024x1536 and 1536x1024.
The AI SDK provider has ib.image() in @inferencebureau/ai-sdk-provider 0.2.0: generateImage generates, and with input images it edits. The models pages show per-image prices and each image model's price or token table.
More in Image edits.