Technical overview
- Input types: Text prompts, up to 4 reference images in the UI (Fal accepts up to 14), JPG/PNG/WebP uploads, and edit instructions—no masks required.
- Outputs: 1K, 2K, or 4K still images with Gemini 3-level detail, clean typography, and accurate brand color.
- Modes: Text-to-image for fresh renders; image-to-image for restyles, relights, and product swaps.
- Pricing: $0.15 per 1K/2K image, $0.30 at 4K. Batch up to 8 shots per run with the same prompt lab and wallet as your video engines.
