Technical Specifications of bria/genfill
| Specification | Details |
|---|---|
| Model ID | bria/genfill |
| Provider | BRIA |
| Model type | Image editing / generative fill model |
| Primary function | Generate, add, modify, or replace objects inside a selected region of an existing image |
| Input modalities | Input image, mask, and text prompt |
| Output modality | Edited image |
| Editing scope | Localized editing within a user-defined masked area |
| Resolution behavior | Returns the edited image at the original resolution without automatic resizing or downscaling |
| Region preservation | Areas outside the supplied mask remain unchanged |
| Transparency handling | Can preserve alpha/transparency when supported and enabled |
| Foundation / underlying stack | Built on BRIA 2.3 with GenFill ControlNet according to BRIA’s product documentation |
| Commercial positioning | Enterprise-oriented and marketed for commercially safer use with licensed training data and controllable outputs |
What is bria/genfill?
bria/genfill is BRIA’s generative fill model for localized image editing. It is designed to let developers generate or replace objects inside a specific masked region of an image using a text prompt, while preserving the rest of the image exactly as-is. BRIA describes GenFill as a production-ready object generation system with strong control, consistency, and enterprise-focused output quality.
In practice, this model is suited for workflows such as inserting products into scenes, replacing image elements, modifying selected objects, or filling erased regions with prompt-guided content. BRIA’s documentation emphasizes that the model works with a supplied mask, is optimized for blob-shaped mask regions, and returns the final image in the original resolution.
A notable part of BRIA’s positioning is commercial readiness. BRIA states that GenFill is built for enterprise use, with controllable outputs and training based on licensed data, making it appealing for teams that need safer deployment in production creative pipelines.
Main features of bria/genfill
- Localized generative editing: Generates content only inside a specified mask, making it useful for targeted object insertion, replacement, and visual modifications.
- Prompt-guided object generation: Uses natural-language prompts to define what should appear in the selected region.
- Pixel-preserving outside the mask: BRIA documents that areas outside the masked region remain unchanged, which is important for deterministic editing workflows.
- Original-resolution output: Returns edited images at the same resolution as the input image rather than automatically downscaling the result.
- Alpha preservation support: Can preserve transparency information when the input includes an alpha channel and the relevant setting is enabled.
- Built for controllability: BRIA positions GenFill as highly controllable and consistent, with its product materials specifically highlighting predictable output and GenFill ControlNet support.
- Enterprise-oriented commercial use: BRIA markets the model for enterprise workflows and emphasizes licensed-data training and commercially safer deployment.
- Useful for production image workflows: Fits applications such as creative tooling, automated content operations, product visualization, and image retouching pipelines. This is an inference based on BRIA’s described API capabilities and product positioning.
How to access and integrate bria/genfill
Step 1: Sign Up for API Key
Sign up on CometAPI and generate your API key from the dashboard. Store the key securely and configure it as an environment variable in your application so your integration can authenticate requests to the bria/genfill API.
Step 2: Send Requests to bria/genfill API
Send HTTPS requests to CometAPI’s unified API endpoint and specify bria/genfill as the model. Include your authentication header plus the required input payload for image editing, such as the source image, mask, and prompt.
curl https://api.cometapi.com/v1/responses \
-H "Authorization: Bearer $COMETAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "bria/genfill",
"input": {
"prompt": "Replace the masked area with a modern wooden chair",
"image_url": "https://example.com/input-image.png",
"mask_url": "https://example.com/input-mask.png"
}
}'
Step 3: Retrieve and Verify Results
Parse the API response and retrieve the generated output from the returned payload. Verify that the edited region matches the intended mask, confirm that unmasked regions remain unchanged, and run any application-level moderation or quality checks required before presenting results to end users.