flux-pro is the image creation entry point for Black Forest Labs' FLUX Pro series, suitable for turning clear text concepts into visual assets that emphasize detail, composition, and style. On this platform, it supports text-to-image generation and can also combine existing images with text instructions for editing. It is suitable for marketing concept visuals, product scene drafts, and illustration exploration, and offers multiple aspect ratios and asynchronous task processing methods.
Clarify capacity, inputs and outputs, and invocation methods before selecting a model.
The FLUX Pro series excels at prompt adherence and visual detail; the dimensions, operations, and result formats above apply to this platform's flux-pro entry point.
Learn what flux-pro can bring to your work.
flux-pro focuses on prompt adherence, visual quality, and image detail. After describing the subject, you can continue specifying the scene, materials, lighting, camera distance, and overall style, turning a piece of text into a clear visual brief. It is suitable for adjusting composition and atmosphere around the same idea, progressively selecting images that better match the goal.
You can create new images from text or submit an existing image link and initiate edits with text instructions. This means the workflow does not need to start from scratch every time: first generate a directional draft, then request changes around the selected image. If the task emphasizes precise local edits and overall consistency, you can further consider the Kontext series.
Square, portrait, and landscape dimensions make it easy to prepare visual drafts for different layouts, and image ratios can also be used to express composition requirements. Generation tasks support specifying a quantity, making it suitable for preparing multiple candidate options at once; combined with asynchronous queries or completion callbacks, image tasks can be integrated into a content production backend without keeping the user interface waiting continuously.
Start with specific tasks to find where the model can be effective.
Enter the campaign theme, main subject, brand color palette, and desired atmosphere to generate candidate images for advertising backgrounds or promotional visuals. Landscape format is suitable for exploring key visual layouts, while portrait format is suitable for content covers. The deliverables can serve as design communication drafts, after which designers can add final copy, logos, and layouts, rather than directly replacing complete ad production.
Generate scene drafts based on descriptions of product appearance, placement, background materials, and lighting, to compare studio-like looks, lifestyle scenes, or artistic expressions. Existing images can also be used as editing inputs. When actual products are involved, results should be compared item by item with the real appearance before deciding which images are suitable for further production.
Turn character settings, environmental relationships, color schemes, and artistic direction into prompts to generate illustration or scene concept candidates. Explore different expressions of the same subject by changing perspective, lighting, and style descriptions. The final deliverable can be a set of direction images for illustrators or art teams to select and refine further, rather than requiring every detail to be completed in one pass.
Choose based on task complexity, input materials, and expected results.
If the goal is to prepare visual assets for a formal project and you care about prompt expression, details, and image quality, flux-pro is worth considering. flux-dev is better suited for development testing and early-stage concept exploration. FLUX.1 [pro] and FLUX1.1 [pro] are different native versions; when calling flux-pro, its name should not be rewritten as 1.1, nor should specific acceleration effects be assumed based on this.
If starting from a text concept and needing multiple composition candidates, use flux-pro; if the core task is context-aware modification of an existing image, it is more suitable to compare flux-kontext-pro or flux-kontext-max. When considering the Flux 2 series, also note differences in size notation: these endpoints use image aspect ratios and cannot directly reuse flux-pro pixel size settings.
From a small-scale task to formal integration.
Clarify the goal, required inputs, and output requirements, using real business examples as a starting point.
Open the trial page, confirm the parameters supported by this endpoint, then submit a small-scale task to review the results.
Keep the complete model ID, use the request format specified in the documentation, and confirm billing rules on the Pricing page.
Before formal use, understand the scope of output quality and capabilities.
Answers to common questions about using flux-pro.
When calling it, use the exact model ID flux-pro; do not rewrite it as FLUX1.1 [pro]. FLUX.1 [pro] and FLUX1.1 [pro] are different versions; when choosing this option, evaluate it based on image quality, prompt adherence, and real-world task results, rather than assuming the speed performance of 1.1.
Yes. Use action=edit, and provide image_url, the edit instruction prompt, and size. Edit tasks do not use count to control quantity. If you want more targeted contextual edits based on an existing image, consider comparing flux-kontext-pro or flux-kontext-max as well.
You can use 1024x1024, 1024x1792, or 1792x1024, or specify an image aspect ratio. First choose square, landscape, or portrait based on the final layout, then describe the subject placement and whitespace requirements in the prompt. When using Flux 2 or Kontext, use their corresponding aspect ratio format instead.
In a generate task, specify the number of images with count; one image is generated by default. The returned data list provides links for each image. This is useful for keeping the main description unchanged, comparing multiple candidates first, and then refining the prompt to continue exploring. count does not apply to edit, and it does not mean every result will be exactly the same.
Submit the model, action, prompt, and size to POST /flux/images. You can set async=true, then query the task result after obtaining task_id; you can also use callback_url to receive completion notifications. After completion, read image_url from data, then pass it to asset management, review, or design workflows.
Model information · Updated: 2026-10-01. For API parameters and billing rules, see the API and pricing sections.
Start with a clear goal and evaluate whether it fits your work based on real results.
This is only a basic call example. See the full docs for more parameters and advanced usage.
View Docs