How it works FLUX Canny Pro?
FLUX Canny Pro is a Black Forest Labs model for image generation where frame structure is defined not just by a prompt, but by an edge map. The Canny detector extracts object boundaries from the source image, and the model adds textures, lighting, and materials on top. This preserves composition, proportions, and object placement while changing style and content—ideal when strict adherence to the prompt's structure is required. In BotHub, the model is accessible without a VPN or foreign card: pay with a local card in rubles only for what you use, with no expiration on funds. With over 250 models available in the same window, you can easily compare results and switch without changing your integration. A unified OpenAI-compatible API and AES-GCM encryption are available if you are embedding generation into your product. It is useful for designers turning sketches into visuals, architects and decorators creating renders from floor plans, marketing teams creating banner series with consistent layouts, and illustrators making storyboards or scene variations.Frequently asked questions about FLUX Canny Pro
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
flux-canny-pro generates images from text descriptions and redraws existing images while preserving their contours and composition. This is useful when you need to change the style, materials, or environment while keeping the scene geometry intact: layouts, interiors, product visualization, and design variations.
You get a raster image as output: the FLUX Pro line is known for detailed, photorealistic results. Specific resolution and file format depend on request parameters, so we do not state fixed values. Request context is 4095 tokens, output limit is 4096.
Step-by-step reasoning is not a feature of flux-canny-pro, and it is not a key function for an image generator: it works in a single pass based on the prompt and reference contours. If you need chain-of-thought reasoning, text models with this mode are available nearby in BotHub.
Function calling and strict JSON output are not supported by this model—its task is to provide an image, not structured text. Integration is handled via the unified OpenAI-compatible BotHub API, so calling it from your code remains familiar and consistent.
Images—yes: the model is designed for them. You upload a reference, contours are extracted, and the new image repeats the original structure. Audio and video inputs are not supported; BotHub has separate models for those formats.
We have no data on prompt languages, so we cannot make promises—test it with your task, it only takes a couple of minutes. What is definitely in Russian is the BotHub interface and support: access from Russia without a VPN or foreign card, and payment in rubles.