Meet Recraft’s New AI Model That Thinks In Design Language

Hey Recrafter!

We are announcing our new image generation model Recraft V3, setting a new standard for excellence in image generation. Over the past week Recraft V3 participated in the Hugging Face's industry-leading Text-to-Image Model Leaderboard by Artificial Analysis. It secured #1 place with ELO rating of 1172.

The new model scores higher than Midjourney, OpenAI, and all other major image generation companies.

Try Recraft V3

What's New and Improved

We are launching several new important features that allow our users to have more control over AI generation: the possibility to specify text size and positions in the image, precise style control, improved inpainting, and new outpainting capabilities:

Recraft V3 is the only model in the world that can generate images with long texts, as opposed to just one or a couple of words. It fits smoothly into your design every time.

Meet the Graphic Design Generator. It is now possible to set the exact positions of text and different elements in the image instead of interpreting the explanation in the prompt.

Drag and drop texts and images into a frame, write the prompt and get the image with those elements positioned the same way as they were placed in the frame. This feature allows you to generate things like posters, ads, and social media materials.

Generate text

Model improvements

Superior prompt understanding and enhanced aesthetics lead to more accurate and appealing visuals.

Outpainting. To do outpainting or a zoom-out, put the image into a frame, write a prompt, and click Recraft.

Improved inpainting. It is now possible to do inpainting on small areas, like adding a phone to a human hand or adjusting facial features in a portrait, giving you more control over small details.