flux-pro-1.0-fill-finetuned

Technical Specifications of flux-pro-1-0-fill-finetuned

flux-pro-1-0-fill-finetuned is CometAPI’s platform identifier for Black Forest Labs’ FLUX.1 Fill [pro] finetune image-editing endpoint. It is designed for inpainting workflows, where you provide an input image plus either a separate mask or an alpha mask, and the model edits only the specified regions based on a text prompt.

Model overview

• Model ID: flux-pro-1-0-fill-finetuned
• Provider / underlying model family: Black Forest Labs FLUX.1 Fill [pro] finetune.
• Primary modality: Image-to-image, image editing, and masked inpainting.
• Commercial use: Listed by CometAPI as supporting commercial use. (cometapi.com)
• CometAPI pricing: $0.096 per request on the CometAPI model page at crawl time. (cometapi.com)

Documented request inputs

• finetune_id: Required ID of the fine-tuned model to apply.
• image: Required base64-encoded source image.
• mask: Optional base64-encoded black-and-white mask; can be omitted if the original image already contains an alpha mask.
• prompt: Text instruction describing the desired edits.
• finetune_strength: Default 1.1, allowed range 0 to 2.0.
• steps: Default 50, allowed range 15 to 50.
• guidance: Default 60, allowed range 1.5 to 100.
• prompt_upsampling: Optional boolean for more creative prompt rewriting.
• seed: Optional integer for reproducibility.
• output_format: jpeg or png, default jpeg.
• safety_tolerance: Integer from 0 to 6, default 2.
• Async delivery support: webhook_url and webhook_secret are supported.

Documented response fields

• id: Generation task ID.
• polling_url: URL for checking task status.
• cost, input_mp, output_mp: Returned usage and image-size-related metadata.

What is flux-pro-1-0-fill-finetuned?

flux-pro-1-0-fill-finetuned is a specialized FLUX image-editing model endpoint for personalized inpainting. In practice, it combines the high-end FLUX.1 Fill [pro] editing workflow with a user-trained finetune, so you can modify selected parts of an existing image while preserving the rest of the scene and maintaining subject, product, brand, or style consistency.

Black Forest Labs describes its FLUX Pro finetuning system as a way to customize FLUX.1 [pro] using your own example images, including specific people, pets, clothing, stickers, brands, or visual styles. The company also states that combining finetuning with FLUX.1 Fill [pro] enables personalized inpainting for iterative editing of a given image.

So, if you already have a fine-tuned FLUX Pro concept model, this endpoint is the editing version that lets you apply that learned concept selectively inside masked regions instead of generating a brand-new image from scratch. That makes it especially useful for ad creative refreshes, product retouching, character-consistent edits, branded asset revisions, and localized image updates. This is an inference based on the endpoint behavior and Black Forest Labs’ finetuning description.

Main features of flux-pro-1-0-fill-finetuned

• Personalized inpainting: Edits only the masked portions of an existing image while using a finetuned concept model to keep the result aligned with your custom subject or style.
• Fine-tune aware editing: Requires a finetune_id, which means the endpoint is built to apply a previously trained FLUX Pro finetune during image modification.
• Mask-based control: Supports both separate masks and alpha-mask workflows, making it practical for precise object replacement, cleanup, enhancement, or region-specific redesign.
• Prompt-guided transformations: Uses natural-language prompts to describe the desired change inside the selected area, enabling flexible visual edits without manual compositing.
• Adjustable style influence: finetune_strength lets you control how strongly the finetuned concept affects the final image, from subtle influence to highly pronounced personalization.
• Quality and creativity controls: Parameters such as steps, guidance, and prompt_upsampling give developers control over fidelity, adherence to prompt, and creative variation.
• Reproducible outputs: Optional seed support helps with repeatable testing and controlled iteration across versions of a creative asset.
• Production-friendly async workflow: The endpoint returns a task ID and polling URL, and it also supports webhooks for automated pipelines.
• Commercial deployment path: CometAPI lists the model as available for commercial use on its platform. (cometapi.com)

How to access and integrate flux-pro-1-0-fill-finetuned

Step 1: Sign Up for API Key

Sign up on CometAPI and create your API key from the dashboard. Once you have your key, you can authenticate requests to the flux-pro-1-0-fill-finetuned API and start integrating the model into your application. CometAPI lists this model as available on its platform with sample code and API resources to help streamline integration. (cometapi.com)

Step 2: Send Requests to flux-pro-1-0-fill-finetuned API

Send a request to the flux-pro-1-0-fill-finetuned API using your preferred HTTP client or SDK. Include your API key, the model ID flux-pro-1-0-fill-finetuned, and the required request payload for your use case. For this model, that typically means providing a source image, a mask or alpha mask, a prompt, and a finetune_id so the endpoint can apply your trained concept during inpainting. Black Forest Labs documents this endpoint as POST /v1/flux-pro-1.0-fill-finetuned with configurable parameters such as finetune_strength, steps, guidance, output_format, and optional webhook settings.

Step 3: Retrieve and Verify Results

Retrieve the generation result using the response metadata returned by the API. For flux-pro-1-0-fill-finetuned, the documented response includes a task id and polling_url, along with usage-related fields such as cost, input_mp, and output_mp. Verify that the edited output matches the intended masked region, preserves surrounding context, and reflects the finetuned concept consistently before promoting it into production workflows.