Now powered by EvoLink gpt-image-2-beta for text-to-image, image-to-image, and editing workflows with strong prompt following and clean iteration.
gpt-image-2-beta
EvoLink async image task
Text
Text to image
Image
Image to image
Edit
Image editing
size: auto + 13 documented aspect ratios
input: prompt + up to 16 reference images
output: async task result, saved to your creations
Preview
Generated images will appear here
Preview
Generated images will appear here
Start from a prompt, then upload references and keep editing in the same playground.
The first version exposes auto plus 13 documented ratios, matching the provider docs instead of over-promising.
GPT Image 2 uses the existing async tasks, credits, polling recovery, history, and storage flow.
This page supports text-to-image and image-to-image generation. Upload a reference image when you want to edit or transform an existing visual.
It supports auto, 1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9, 9:21, 2:1, 1:2, 3:1, and 1:3.
Yes. GPT Image 2 uses the same history and resource storage flow as the other production image models.