Feature Requests

Please add Ideogram v3 as an external image generator. That would be a huge value added move for Recraft and fill a gap in the styles with quality output. I really did not like the output of previous versions of Ideogram, but v3 is crushing it!

All of the Illustrations we create are only rendered at 96dpi, irrespective of the size of the image we create. This makes it too low for good printing. We can of course upscale or upgrade it but it still renders at 96dpi. Whilst we can create an SVG, the styling comes out better in the Illustration mode. So, please, can you give us the option to choose 300 DPI when creating larger images for printing?

Would be very happy with the ability to highlight multiple images and then have the option to either upscale, or upscale and then creative upscale and it put in a queue so we can come back when it is complete with a zip download, as it can be time consuming waiting around

Close-Up, Side Profile, Extreme Close-Up, Medium Close-Up, Medium, Rear-View Over-The-Shoulder, From Behind, Three-Quarter, Eye Level, Hero, Low Angle, Ground Level, Aerial, Over The Top, Top-Down, High Angle

Apparently your image system was trained only on vanishing point views, and cannot create orthographic views. My design work needs orthographic objects, which are horizontal, equidistant left and right ends. NOT just perspective, vanishing point. Here’s the difference. https://photos.app.goo.gl/TcuZoku7GWsyG7AQA

Hello, I want to download the image set section as SVG in bulk, not individually.

Is it actually possible to use external models as styles via API? I have not found anything in the documentation.

I’d love to see Recraft support structure-guided generation using control inputs like Canny edge maps, human poses, depth maps, or segmentation masks. This would provide much finer control over composition and layout, similar to techniques used in ControlNet and IP-Adapter in the ML community. It would significantly improve precision in generating consistent characters, poses, or environments, especially in professional workflows. Allowing users to upload or generate these structures directly within Recraft would be a game-changer.

I've noticed that Recraft AI struggles with generating black-and-white ink art compared to colored pictures. It would be great if the AI could be improved to handle black-and-white ink art better, providing more consistent and high-quality results.

I am creating mockup using this tutorial: https://www.recraft.ai/mockup-generator Expected result: I am able to drag a png with my screen on the generated in recraft mockup and adjust it with persepctive / WRAPPING toll (similar to phonoshop) to make it look like real screen on device. I am using paid version to generate multiple mockups this way, because it supposed to be much faster then in photoshop itself. Current result: you can see on the screenshots. I didnt find any way how I can overlay my screen, so I have to go back to photoshop-like tools. No hotkeys or anything to fix that. And device size doesnt matter. Other than that - navigation in choosing model/style was changed and it is very uncomfortable to explore them. Please at least make search working without "Enter"