Shot by Text - Original
Describe the scene, keep the product shot
- image
- output_image
ShotByTextOriginal (shown as Shot by Text - Original) is the text-driven member of Bria's product-shot family. Instead of giving it a reference image to copy, you hand it a scene_description - "minimalist bathroom, morning light, marble shelf" - and it builds that scene around your product photo, keeping the original composition of the source shot.
Where ShotByImage reads a photo's vibe, this reads your words. It's the right node when you have a specific scene in mind that no reference photo quite captures, or when you're generating a whole catalog and want every shot to follow the same written art direction.
How it works
Two required inputs: image (your product) and scene_description (what the scene should look like). It POSTs to Bria's /v1/product/lifestyle_shot_by_text endpoint, powered by BRIA's ControlNet background-generation. The "Original" placement means no repositioning - Bria keeps your product photo's composition and rebuilds the environment to match your description.
The text-specific controls are where this node gets interesting:
- mode -
base,fast, orhigh_control, defaultfast. A three-way speed/control tradeoff:fastis the default workhorse,baseis the plain pipeline,high_controlis the slower, more prompt-faithful setting. Usehigh_controlfor the hero shots,fastfor iterating. - optimize_description (default true) - Bria rewrites/expands your scene description internally before generating. Keep it on for vague prompts; turn it off when you've written exact wording and don't want it paraphrased.
- exclude_elements - a plain-text list of things you don't want in the scene ("no faucet, no towels"), sent to the API explicitly. Genuinely useful for keeping products from inheriting clutter.
- Plus the shared toggles: force_rmbg, content_moderation.
One output: output_image (IMAGE).
Installing it
Part of the BRIA AI API nodes pack (Bria-AI/ComfyUI-BRIA-API). ComfyUI Manager → search BRIA API → install, or:
cd ComfyUI/custom_nodes
git clone https://github.com/Bria-AI/ComfyUI-BRIA-API
Restart ComfyUI; it's under API Nodes. No downloads, no GPU - cloud call, needs a BRIA API token from platform.bria.ai, billed per generation.
Practical notes
Prompt writing still matters, just in a more structured way: concrete nouns and scene words ("marble", "matte black", "dappled window light") beat adjectives ("nice", "aesthetic"). And because "Original" locks the composition, this node is best for shots where the product's own framing is already right - when you want Bria to recompose the whole thing, the Automatic, Manual Placement, or Custom Coordinates variants take over.
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| api_key | STRING | BRIA_API_TOKEN | — |
| image | IMAGE | — | |
| scene_description | STRING | — | |
| mode | COMBO | fast | 3 options: base, fast, high_control |
| force_rmbgopt | BOOLEAN | false | — |
| content_moderationopt | BOOLEAN | false | — |
| optimize_descriptionopt | BOOLEAN | true | — |
| exclude_elementsopt | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| output_image | IMAGE | — |