FLUX.1.1 [pro] Ultra
Nano Banana, as a ComfyUI node
- advancedFeatures.watermark.image
- image
You've seen the banana. "Nano Banana" - the community name for FLUX.1.1 [pro] Ultra - is the fastest, most photorealistic image model Black Forest Labs has ever shipped, and it's closed-weights, API-only, and genuinely impressive at real-world photography. Faces, skin, materials, lighting: it reads as a photo in a way that the open FLUX tiers never quite managed. The "Nano" part is the speed (it's fast - sub-second class in good conditions), and the "Banana" part is the banana. This node is that API as a ComfyUI node, via Runware.
It's the natural upgrade path from the plain [pro] node: same family, same wiring, better photorealism and a couple of extra dials.
How it works
A thin taskType: imageInference REST call with model AIR bfl:2@2. Runware runs the closed BFL model and hands back an IMAGE tensor. All the smarts are proprietary; your job is just prompt, size, and the two BFL-specific toggles.
Inputs that matter
- positivePrompt - required. Write it like a photograph you're describing to a very literal assistant.
- width / height - default 1024; no stated range in the schema, and Ultra handles high resolution natively, so feel free to go big.
- providerSettings.bfl.raw - the interesting one. On, you get the direct model output with no post-processing. Off, BFL's polishing pipeline smooths things for that extra "finished photo" feel. People who like crisp, filmic realism tend to like raw on; people who want the glossy default leave it off.
- providerSettings.bfl.promptUpsampling - prompt expansion, off by default.
- providerSettings.bfl.safetyTolerance - 0–5, default 2; the moderation dial, same as the rest of the BFL pro family.
- seed / numberResults - reproducibility and cheap variations.
Install
cd ComfyUI/custom_nodes
git clone https://github.com/Runware/ComfyUI-Runware
pip install -r ComfyUI-Runware/requirements.txt
Or ComfyUI Manager → search Runware → Install → restart. API key in ComfyUI Settings → Runware API key or RUNWARE_API_KEY.
Gotchas
- It's the premium tier, so watch the per-run cost. The title bar shows it - use it to keep track before a long batch surprises you.
- Moderation is BFL-side and adjustable only through
safetyTolerance. - Photorealism this good is a double-edged sword: it'll reproduce real people's likenesses convincingly, and BFL's moderation is exactly why. Don't be that person.
If "it looks like a photo" is the actual requirement, this is the node in the pack that delivers it fastest. The raw toggle is the one setting worth experimenting with - it changes the character of the output more than anything else on the node.
Inputs (21)
| Name | Type | Default | Description |
|---|---|---|---|
| positivePrompt | STRING | Text prompt describing elements to include in the generated output. | |
| advancedFeatures.watermark.imageopt | IMAGE | — | |
| sizeopt | COMBO | 2K (1:1) | Output size. Only one of resolution or dimensions is sent. |
| seedopt | INT | 00–4294967295 | Random seed for reproducible generation. When not provided, a random seed is generated in the unsigned 32-bit range. |
| numberResultsopt | INT | 11–20 | Number of results to generate. Each result uses a different seed, producing variations of the same parameters. |
| advancedFeaturesopt | BOOLEAN | false | Enable to set advancedFeatures. Off uses the model's default. |
| advancedFeatures.watermark.bgColoropt | STRING | Background color in hex format. | |
| advancedFeatures.watermark.displayPositionopt | COMBO | (default) | Watermark position. |
| advancedFeatures.watermark.fontColoropt | STRING | Text color in hex format. | |
| advancedFeatures.watermark.opacityopt | BOOLEAN | false | Enable to set advancedFeatures.watermark.opacity. Off uses the model's default. |
| advancedFeatures.watermark.opacity_valueopt | FLOAT | 0.100.1–1 | Watermark opacity. |
| advancedFeatures.watermark.textopt | STRING | Watermark text. | |
| providerSettings.bfl.promptUpsamplingopt | BOOLEAN | false | Enable automatic enhancement and expansion of the input prompt. |
| providerSettings.bfl.rawopt | BOOLEAN | false | Enable raw mode to produce images closer to direct model output without post-processing. |
| safetyopt | BOOLEAN | false | Enable to set safety. Off uses the model's default. |
| safety.checkContentopt | BOOLEAN | false | Enable or disable content safety checking. |
| providerSettings.bfl.safetyToleranceopt | INT | 20–5 | Tolerance level for content moderation. Lower values are stricter; higher values are more permissive. |
| ttlopt | BOOLEAN | false | Enable to set ttl. Off uses the model's default. |
| ttl_valueopt | INT | 60 | Time-to-live (TTL) in seconds for generated content. Only applies when `outputType` is `URL`. |
| outputFormatopt | COMBO | JPG | File format for the generated image. |
| outputQualityopt | INT | 9520–99 | Compression quality of the output. Higher values preserve quality but increase file size. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |