SF WaveSpeed Nano Banana Pro Edit Multi
Two Nano Banana Pro edits per run from the same prompt
- client
- output_images
The regular Nano Banana Pro Edit node gives you one edited image per run. SF WaveSpeed Nano Banana Pro Edit Multi gives you two - same prompt, same references, two different results. It's the "give me options" node. When you've got the edit instruction right but you're not sure which of two interpretations will look better, this is how you find out without running the single-edit node twice (and paying twice for the same prompt).
How it works
It's the sibling of the single-edit node with one structural difference: num_images is fixed at 2, and the endpoint is google/nano-banana-pro/edit-multi. You feed the prompt and up to six image URLs, the API generates two edited variants from that same input, and the node returns both. Everything else - aspect ratio, output format, sync mode - behaves like the single-edit version. The output is output_images (IMAGE), which in ComfyUI means a batch of two images, ready for the usual batch-splitting or comparison tricks if you want to view them side by side.
The inputs that matter
- client - from SF WaveSpeed Client.
- prompt - the edit instruction. Since two results come from this one string, make it specific; the model's two interpretations are where the options live, not in a vague prompt.
- image_1 … image_6 - public URLs,
image_1required. - aspect_ratio - default 3:2 here (the single node defaults to 1:1) - a slightly wider canvas by default.
- num_images - present but fixed at 2; don't expect to change it.
- output_format, enable_sync_mode - as usual.
Output: output_images (IMAGE batch of 2).
Installing it
In the SF ComfyUI Nodes pack from Stillfront. ComfyUI Manager → "SF ComfyUI Nodes" → Install → restart, or:
cd ComfyUI/custom_nodes
git clone https://github.com/Stillfront/comfyui-sf-nodes.git
cd comfyui-sf-nodes
pip install -r requirements.txt
Plus a WaveSpeed API key through SF WaveSpeed Client.
Common issues
The public-URL rule for input images applies exactly as it does everywhere else in this family. And a small expectation check: two images out means roughly double the generation cost of a single edit, so this node shines for picking between options, not for spamming identical outputs. If you find the two variants are coming out nearly identical, that's usually a prompt that's too prescriptive - loosen it slightly and the second interpretation will actually differ. Also, don't forget the output is a batch: a single preview node may only show you the first frame, so split the batch if you want to eyeball both edits side by side.
Inputs (12)
| Name | Type | Default | Description |
|---|---|---|---|
| client | WAVESPEED_AI_API_CLIENT | — | |
| prompt | STRING | Text description of the edit to perform | |
| image_1 | STRING | Primary input image URL (connect from Upload Image node) | |
| image_2 | STRING | Second input image URL (optional) | |
| image_3 | STRING | Third input image URL (optional) | |
| image_4 | STRING | Fourth input image URL (optional) | |
| image_5 | STRING | Fifth input image URL (optional) | |
| image_6 | STRING | Sixth input image URL (optional) | |
| aspect_ratio | COMBO | 3:2 | Aspect ratio of the output images |
| num_images | INT | 22–2 | Number of edited images to generate (fixed at 2) |
| output_format | COMBO | png | Output format - use PNG for transparency support |
| enable_sync_mode | BOOLEAN | true | Wait for generation to complete before returning |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| output_images | IMAGE | — |