new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

AI Asset Pipeline · all subjects

api/image-generation

4 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Submit image generation request endpoint

POST /v1/images/generations submits an asynchronous image generation request. The request body must include: model (string, required), prompt (string, required), size (aspect ratio string, required), n (integer 1–4, required), resolution (string, required), image_urls (array, optional, for i2i/edit), and official_fallback (boolean, optional). The response returns a task_id for polling.

Supported aspect ratios for image generation

The size parameter accepts: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9. Additional ratios 1:4, 4:1, 1:8, 8:1 are available on gemini-3.1. The gpt-image-2 model also accepts WxH format (width×height in pixels).

Image URL limits for i2i and edit operations

The image_urls parameter accepts up to 14 images per request, with each image limited to 10 MB maximum. Images can be submitted as URLs or base64 data URIs for image-to-image or edit workflows.

Number of images per request by model

The n parameter controls images per request. Most models support n=1–4. gpt-image-2 supports n=1 only. seedream models support n up to 15.

Give your agent this brain