What Seedream 5.0 Flash is good at
- Flat rate at every resolution: One price per generated image at 1K, 1.5K, and 2K, so the resolution you pick does not change the cost
- Generation and editing in one node: Generate from a prompt, or attach a reference image and describe the change in a sentence
- Multi-reference input: Accepts up to 10 reference images, which is enough to combine several subjects, products, or styles in one composition
- Text rendering: Handles captions, labels, and layout text inside the generated image
- Single image per run: Each run returns one image. The batch and sequential generation options available on other Seedream models are not exposed for this model
- Layer separation: Selectable as the model in the layer separation node for turning a flat image into an editable layer stack
Example outputs
Text-to-image generation, producing one image at the selected size preset:
Image editing from a single reference image and a text instruction:
Seedream 5.0 Flash text-to-image workflow
Generate an image from a text prompt. The workflow returns a single image at the size preset selected on the node.
Run on Comfy Cloud
Open in Comfy Cloud
Download Workflow
Download JSON or search “Seedream 5.0 Flash” in Template Library
Seedream 5.0 Flash image edit workflow
Edit an existing image with a text instruction. Connect the source image to the node’simage_1 input and describe the change you want, from swapping a background to rebuilding a product set around the subject in the reference.
Run on Comfy Cloud
Open in Comfy Cloud
Download Workflow
Download JSON or search “Seedream 5.0 Flash” in Template Library
Get started
- Update ComfyUI to version 0.37.3 or later
- Go to Template Library, search for
Seedream 5.0 Flash