Figma: Generate Multiple Images From One Prompt, or Compare Up to Three Models Side by Side

FigmaView original changelog

Figma now lets designers generate several image options at once when using Make image or Edit image. A single model can return 2, 4 or 6 images from one prompt, or the same prompt can run across up to three image models at once, with one result per model. The feature is available in Figma Design, FigJam, Figma Slides and Figma Draw, so users can compare interpretations and keep the best one.

Key Takeaways

  • A single prompt can now return 2, 4 or 6 images, so designers see a range of options instead of re-running the same request.
  • The same prompt can run across up to three image models at once, returning one image per model for direct comparison.
  • Place all on canvas lays every result side by side, which speeds up choosing a winner.
  • The feature works in Figma Design, FigJam, Figma Slides and Figma Draw, not just in the design editor.
  • It is listed for all plans, but Starter users get AI image tools only in Figma Design.
  • Generation uses the shared Figma AI credit system, and the per-image credit cost is not stated in the help article.

Several Image Options From One Prompt

Figma added a way to generate more than one image per request when using the AI image tools Make image and Edit image. Until now, reaching the right image often meant re-running a prompt and hoping for a better result. With this release, Figma lets users ask for several candidates in a single step and pick the one that fits the design. The update is listed under Assets and applies to Figma Design, FigJam, Figma Slides and Figma Draw.

Two Ways to Generate

One model, multiple images

When a single image model is selected, the model picker now offers a choice of 2, 4 or 6 images per prompt. This suits cases where a prompt is already well defined and the goal is to see a range of variations on the same idea, such as different compositions of a hero image or alternate illustrations for a slide.

Multiple models, one image each

Alternatively, users can add more models to the same request. Figma runs the prompt across up to three image models at the same time and returns one image per selected model. This makes it possible to compare how different models interpret the same wording and keep the result that looks best, without switching tools or re-entering the prompt.

Comparing Results on the Canvas

When filling an existing layer, users can preview each option in place. A Place all on canvas action puts every result on the canvas next to each other so they can be compared side by side before one is chosen. Selecting where the image should go is the first step: pick an existing layer to fill, or select nothing to create a new layer.

How to Use It

Open the Actions menu and choose Make an image (in Figma Buzz, use the inline toolbar's Choose media). Enter a detailed prompt, select one or more models, optionally add a reference image, then click Send or press Command Enter. The file needs edit access.

Availability and Credits

Figma lists the feature for all plans. On the Starter plan, AI image tools are available only in Figma Design. Image generation draws on the credit system shared across all Figma AI features and products. Figma's article does not state how many credits each image costs, so users should check the linked credits documentation before generating large batches.