Overview
Bria’s Image Generation API provides builders with a flexible and controllable way to generate high-quality visuals from textual prompts or reference images.
It supports various resolutions, aspect ratios, and generation speeds, enabling scalable and efficient production of photos, illustrations, and vector graphics for commercial use.
Advanced Customization and Access:
As part of Bria’s Source Code & Weights product, developers seeking deeper customization can access Bria’s source-available GenAI models via Hugging Face.
This allows full control over fine-tuning, pipeline creation, and integration into proprietary workflows—empowering AI teams to develop and optimize their own generative AI solutions.
Some of the APIs below support various guidance methods to provide greater control over generation. These methods enable to guide the generation using not only a textual prompt, but also visuals.
The following APIs support guidance methods:
/text-to-image/base/text-to-image/fast/text-to-vector/base/text-to-vector/fast
ControlNets:
A set of methods that allow conditioning the model on additional inputs, providing detailed control over image generation.
- controlnet_canny: Uses edge information from the input image to guide generation based on structural outlines.
- controlnet_depth: Derives depth information to influence spatial arrangement in the generated image.
- controlnet_recoloring: Uses a grayscale version of the input image to guide recoloring while preserving geometry.
- controlnet_color_grid: Extracts a 16x16 color grid from the input image to guide the color scheme of the generated image.
You can specify up to two ControlNet guidance methods in a single request. Each method requires an accompanying image and a scale parameter to determine its impact on the generation inference.
When using multiple ControlNets, all input images must have the same aspect ratio, which will determine the aspect ratio of the generated results.
To use ControlNets, include the following parameters in your request:
guidance_method_X: Specify the guidance method (where X is 1, 2). If the parameterguidance_method_2is used,guidance_method_1must also be used. If you want to use only one method, useguidance_method_1.guidance_method_X_scale: Set the impact of the guidance (0.0 to 1.0).guidance_method_X_image_file: Provide the base64-encoded input image.
| Guidance Method | Prompt | Scale | Input Image | Guidance Image | Output Image |
|---|---|---|---|---|---|
| ControlNet Canny | An exotic colorful shell on the beach | 1.0 |  |  |  |
| ControlNet Depth | A dog, exploring an alien planet | 0.8 |  |  |  |
| ControlNet Recoloring | A vibrant photo of a woman | 1.00 |  |  |  |
| ControlNet Color Grid | A dynamic fantasy illustration of an erupting volcano | 0.7 |  |  |  |
Image Prompt Adapter:
This method offers two modes:
- regular: Uses the image’s content, style elements, and color palette to guide generation.
- style_only: Uses the image’s high-level style elements and color palette to influence the generated output.
To use Image Prompt Adapter as guidance, include the following parameters in your request:
image_prompt_mode: Specify how the input image influences the generation.image_prompt_scale: Set the impact of the provided image on the generated result (0.0 to 1.0).image_prompt_file: Provide the base64-encoded image file to be used as guidance.
or
image_prompt_urls: Provide a list of URLs pointing to publicly accessible images to be used as guidance.
| Guidance Method | Prompt | Mode | Scale | Guidance Image | Output Image |
|---|---|---|---|---|---|
| Image Prompt Adapter | A drawing of a lion laid on a table. | regular | 0.85 |  |  |
| Image Prompt Adapter | A drawing of a bird. | style | 1 |  |  |
https://engine.prod.bria-api.com/v1/
Request
Description
This Image Generation pipeline enables builders to integrate the generation of high-quality, photorealistic or artistic images into their products. It harnesses the power of Bria’s proprietary text-to-image foundation models, enhanced with additional capabilities that optimize output quality and provide greater control. The pipeline supports native generation at 1MP resolution and accommodates a wide range of aspect ratios to fit diverse needs.
Examples:
prompt: A professional headshot of a CEO
.png)
.png)
.png)
.png)
This API endpoint supports content moderation via an optional parameter that can prevent generation if input images contain inappropriate content, and filters out unsafe generated images - the first blocked input image will fail the entire request.
The prompt you would like to use to generate images. Bria currently supports prompts in English only, excluding special characters.
How many images you would like to generate. When using any Guidance Method, please use the value 1.
The aspect ratio of the image. When a ControlNet is being used, the aspect ratio is defined by the guidance image and this parameter is ignored.
Determines the response mode. When true, responses are synchronous. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt, model type and model version. You can exclude this parameter if you are not interested in recreating your results. This parameter is optional.
Specify here elements that you didn't ask in the prompt, but are being generated, and you would like to exclude. This parameter is optional. Bria currently supports prompts in English only.
The number of iterations the model goes through to refine the generated image. This parameter is optional.
Determines how closely the generated image should adhere to the input text description. This parameter is optional.
Which medium should be included in your generated images. This parameter is optional.
When set to true, enhances the provided prompt by generating additional, more descriptive variations, resulting in more diverse and creative output images. Note that turning this flag on may result in a few additional seconds to the inference time. Built with Meta Llama 3.
When set to true, generates images with richer details, sharper textures, and enhanced clarity.
Slightly increases generation time per image.
When enabled (default: true), the input prompt is scanned for NSFW or ethically restricted terms before image generation. If the prompt violates Bria's ethical guidelines, the request will be rejected with a 408 error.
When enabled, applies content moderation to both input visuals and generated outputs.
For input images:
- Processing stops at the first image that fails moderation
- Returns a 422 error with details about which parameter failed
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
Flags prompts with potential IP content. If detected, a warning will be included in the response.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. The support for guidance methods is currently available only when model_version="2.3". This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_1. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. The support for guidance methods is currently available only when model_version="2.3". This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_2. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
regular: Uses the image’s content, style elements, and color palette to guide generation.style_only: Uses the image’s high-level style elements and color palette to influence the generated output. The support for image prompt is currently available only when model_version="2.3".
The image file to be used as guidance, in base64 format. Accepted formats are jpeg, jpg, png, webp. The support for image prompt is currently available only when model_version="2.3". Maximum file size 12MB. This image can be of any aspect ratio, even when it's not alligned with the one defined in the parameter 'aspect_ratio' or by visuals provided to the ControlNets.
A list of URLs of images that should be used as guidance. The images can be of different aspect ratios. Accepted formats are jpeg, jpg, png, webp. The URLs should point to accessible, publicly available images. The support for image prompt is currently available only when model_version="2.3".
https://engine.prod.bria-api.com/v1/text-to-image/base/{model_version}
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
'https://engine.prod.bria-api.com/v1/text-to-image/base/{model_version}' \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "a book",
"num_results": 2,
"sync": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
This Image Generation pipeline is optimized for speed, enabling builders to rapidly generate high-quality, photorealistic or artistic images without compromising visual fidelity. It leverages Bria’s proprietary text-to-image foundation models, combined with performance-oriented enhancements that ensure faster generation while maintaining control and flexibility. The pipeline supports native generation at 1MP resolution and accommodates a wide range of aspect ratios, making it ideal for applications that require quick turnaround with consistent output control.
An example:
prompt: A portrait of a Beautiful and playful ethereal singer, art deco, fantasy, intricate art deco golden designs, elegant, highly detailed, sharp focus, blurry background, teal and orange shades
BRIA FAST model 2.3:
.png)
.png)
This API endpoint supports content moderation via an optional parameter that can prevent generation if input images contain inappropriate content, and filters out unsafe generated images - the first blocked input image will fail the entire request.
The prompt you would like to use to generate images. Bria currently supports prompts in English only, excluding special characters.
The aspect ratio of the image. The aspect ratio of the image. When a ControlNet is being used, the aspect ratio is defined by the guidance image and this parameter is ignored.
Determines the response mode. When true, responses are synchronous. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt, model type and model version. You can exclude this parameter if you are not interested in recreating your results. This parameter is optional.
The number of iterations the model goes through to refine the generated image. This parameter is optional.
Which medium should be included in your generated images. This parameter is optional.
When set to true, enhances the provided prompt by generating additional, more descriptive variations, resulting in more diverse and creative output images. Note that turning this flag on may result in a few additional seconds to the inference time.
When set to true, generates images with richer details, sharper textures, and enhanced clarity.
Slightly increases generation time per image.
When enabled (default: true), the input prompt is scanned for NSFW or ethically restricted terms before image generation. If the prompt violates Bria's ethical guidelines, the request will be rejected with a 408 error.
When enabled, applies content moderation to both input visuals and generated outputs.
For input images:
- Processing stops at the first image that fails moderation
- Returns a 422 error with details about which parameter failed
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
Flags prompts with potential IP content. If detected, a warning will be included in the response.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_1. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_2. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
regular: Uses the image’s content, style elements, and color palette to guide generation.style_only: Uses the image’s high-level style elements and color palette to influence the generated output.
The image file to be used as guidance, in base64 format. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. This image can be of any aspect ratio, even when it's not alligned with the one defined in the parameter 'aspect_ratio' or by visuals provided to the ControlNets.
A list of URLs of images that should be used as guidance. The images can be of different aspect ratios. Accepted formats are jpeg, jpg, png, webp. The URLs should point to accessible, publicly available images.
https://engine.prod.bria-api.com/v1/text-to-image/fast/{model_version}
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
'https://engine.prod.bria-api.com/v1/text-to-image/fast/{model_version}' \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "a book",
"num_results": 2,
"sync": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
This Image Generation pipeline is designed for builders working on projects that demand maximum image detail and clarity. It leverages Bria’s proprietary text-to-image foundation models, enhanced to deliver high-resolution outputs while preserving control and flexibility. The pipeline supports native generation at 1920×1080 resolution (or 1536×1536 for square aspect ratios) and accommodates a wide range of aspect ratios, making it ideal for use cases where visual precision and output quality are critical.
Examples:
prompt: A photo of detailed short female blond hair viewed from behind, with rich texture and clearly visible individual strands that give depth and realism, and featuring subtle waves reflect light
BRIA HD model 2.2:
prompt: A portrait of a Beautiful and playful ethereal singer, art deco, fantasy, intricate art deco golden designs, elegant, highly detailed, sharp focus, blurry background, teal and orange shades
BRIA HD model 2.2:
.png)
.png)
This API endpoint supports content moderation via an optional parameter that filters out unsafe generated images.
The prompt you would like to use to generate images. Bria currently supports prompts in English only, excluding special characters.
The aspect ratio of the image.
Determines the response mode. When true, responses are synchronous, and applicable to single-image requests. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt, model type and model version. You can exclude this parameter if you are not interested in recreating your results. This parameter is optional.
Specify here elements that you didn't ask in the prompt, but are being generated, and you would like to exclude. This parameter is optional. Bria currently supports prompts in English only.
The number of iterations the model goes through to refine the generated image. This parameter is optional.
Determines how closely the generated image should adhere to the input text description. This parameter is optional.
Which medium should be included in your generated images. This parameter is optional.
When set to true, enhances the provided prompt by generating additional, more descriptive variations, resulting in more diverse and creative output images. Note that turning this flag on may result in a few additional seconds to the inference time.
When set to true, generates images with richer details, sharper textures, and enhanced clarity.
Slightly increases generation time per image.
When enabled, applies content moderation to generated outputs.
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
https://engine.prod.bria-api.com/v1/text-to-image/hd/{model_version}
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
'https://engine.prod.bria-api.com/v1/text-to-image/hd/{model_version}' \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "a book",
"num_results": 2,
"sync": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
This Vector Generation pipeline enables builders to integrate the generation of high-quality, editable vector graphic assets into their products or workflows. It leverages Bria’s proprietary text-to-image foundation models, enhanced to produce vector-compatible outputs while maintaining control and flexibility. The pipeline supports generation of scalable assets such as icons, logos, and illustrations, making it ideal for use cases requiring vector graphics.
Examples:
prompt: A sticker of a cute kitten
prompt: A beautiful butterfly
On the left, a generated vector illustration. On the right, the same illustration after being re-colored in a vector editor
This API endpoint supports content moderation via an optional parameter that can prevent generation if input images contain inappropriate content, and filters out unsafe generated images - the first blocked input image will fail the entire request.
The prompt you would like to use to generate images. Bria currently supports prompts in English only, excluding special characters.
The aspect ratio of the image. When a ControlNet is being used, the aspect ratio is defined by the guidance image and this parameter is ignored.
The aspect ratio of the image.
Determines the response mode. When true, responses are synchronous. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt, model type and model version. You can exclude this parameter if you are not interested in recreating your results. This parameter is optional.
Specify here elements that you didn't ask in the prompt, but are being generated, and you would like to exclude. This parameter is optional. Bria currently supports prompts in English only.
The number of iterations the model goes through to refine the generated image. This parameter is optional.
Determines how closely the generated image should adhere to the input text description. This parameter is optional.
When enabled (default: true), the input prompt is scanned for NSFW or ethically restricted terms before image generation. If the prompt violates Bria's ethical guidelines, the request will be rejected with a 408 error.
When enabled, applies content moderation to both input visuals and generated outputs.
For input images:
- Processing stops at the first image that fails moderation
- Returns a 422 error with details about which parameter failed
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
Flags prompts with potential IP content. If detected, a warning will be included in the response.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_1. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_2. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
regular: Uses the image’s content, style elements, and color palette to guide generation.style_only: Uses the image’s high-level style elements and color palette to influence the generated output.
The image file to be used as guidance, in base64 format. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. This image can be of any aspect ratio, even when it's not alligned with the one defined in the parameter 'aspect_ratio' or by visuals provided to the ControlNets.
A list of URLs of images that should be used as guidance. The images can be of different aspect ratios. Accepted formats are jpeg, jpg, png, webp. The URLs should point to accessible, publicly available images.
https://engine.prod.bria-api.com/v1/text-to-vector/base/{model_version}
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
'https://engine.prod.bria-api.com/v1/text-to-vector/base/{model_version}' \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "A simplistic flat icon of a megaphone",
"num_results": 2,
"sync": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
This Vector Generation pipeline is optimized for speed, enabling builders to quickly integrate the generation of high-quality, editable vector graphic assets into their products or workflows. It leverages Bria’s proprietary text-to-image foundation models, enhanced to accelerate output while maintaining control and flexibility. The pipeline supports generation of scalable assets such as icons, logos, and illustrations, making it ideal for use cases that require rapid delivery of vector graphics.
Examples:
prompt: An icon of a bird with a blue head and yellow beak against a solid background
This API endpoint supports content moderation via an optional parameter that can prevent generation if input images contain inappropriate content, and filters out unsafe generated images - the first blocked input image will fail the entire request.
The prompt you would like to use to generate images. Bria currently supports prompts in English only, excluding special characters.
The aspect ratio of the image. When a ControlNet is being used, the aspect ratio is defined by the guidance image and this parameter is ignored.
Determines the response mode. When true, responses are synchronous. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt, model type and model version. You can exclude this parameter if you are not interested in recreating your results. This parameter is optional.
The number of iterations the model goes through to refine the generated image. This parameter is optional.
When enabled (default: true), the input prompt is scanned for NSFW or ethically restricted terms before image generation. If the prompt violates Bria's ethical guidelines, the request will be rejected with a 408 error.
When enabled, applies content moderation to both input visuals and generated outputs.
For input images:
- Processing stops at the first image that fails moderation
- Returns a 422 error with details about which parameter failed
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
Flags prompts with potential IP content. If detected, a warning will be included in the response.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_1. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
Which guidance type you would like to include in the generation. Up to 2 guidance methods can be combined during a single inference. This parameter is optional.
The image that should be used as guidance, in base64 format, with the method defined in guidance_method_2. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. If more then one guidance method is used, all guidance images must be of the same aspect ratio, and this will be the aspect ratio of the generated results. If guidance_method_1 is selected, an image must be provided.
regular: Uses the image’s content, style elements, and color palette to guide generation.style_only: Uses the image’s high-level style elements and color palette to influence the generated output.
The image file to be used as guidance, in base64 format. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB. This image can be of any aspect ratio, even when it's not alligned with the one defined in the parameter 'aspect_ratio' or by visuals provided to the ControlNets.
A list of URLs of images that should be used as guidance. The images can be of different aspect ratios. Accepted formats are jpeg, jpg, png, webp. The URLs should point to accessible, publicly available images.
https://engine.prod.bria-api.com/v1/text-to-vector/fast/{model_version}
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
'https://engine.prod.bria-api.com/v1/text-to-vector/fast/{model_version}' \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "A simplistic flat icon of a megaphone",
"num_results": 2,
"sync": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
The /reimagine endpoint in Bria’s API allows guiding image generation not just with prompts but also by using an input image. This feature retains the original structure and depth of the input while incorporating new materials, colors, and textures to create fresh visuals.
Key Benefits
Simplified Structure Guidance: Use a reference image to replicate its outline and depth, reducing the need for complex prompts and minimizing trial and error.
Versatile Input/Output Pairings:
- Convert illustrations, sketches, or photos into new illustrative outputs.
- Transform photos into variations that maintain the original layout.
Adjustable Structure Influence: Control how much the input image's structure impacts the output on a scale from 0 to 1, allowing for diverse creative results.
Aspect Ratio Preservation: Ensures the output maintains the reference image's aspect ratio for layout consistency. The output resolution is approximately 1 megapixel.
Seamless Integration with Tailored Generation: Combine structural references with tailored models to include unique IP characteristics in the generated outputs.
Potential Use Cases
Enhanced Creative Control for Platforms & Editing Tools
Empower creative platforms and editing tools with advanced levels of control and flexibility for generating visual content.
Maintain Spatial Consistency
Structure reference image
+(1).png)
Generated Visual (combined into a gif)
.gif)
Convert Sketches to Illustrations
Structure reference image
.jpg)
prompt: A watercolor painting of a lively urban street featuring a red vintage car parked in front of multi-story buildings, where soft, fluid brushstrokes capture the subtle gradients in the building facades, with warm earth tones blending into cool blues and grays for the shadows, giving the scene a nostalgic and dreamy atmosphere.
structure_ref_influence: 0.75
Generated Visual
.png)
Generate Diverse Variations
Structure reference image
+(1).jpg)
prompt: A ginger kitten sits on a textured beige surface, surrounded by soft balls of yarn.
structure_ref_influence: 0.75
Generated Visual
Stylize Typography and Logos
Structure reference image
+(1).png)
prompt: curled orange peel.
structure_ref_influence: 0.1
Generated Visual
.png)
Interoperability with Tailored Generation
Reskin Gaming Assets: Maintain the structure and detail of assets while updating textures and colors for fresh looks without altering the original shape or layout.
Iterate on Gaming Asset Designs: Simplifies design iteration for gaming assets, enabling rapid exploration and refinement.
Customize Marketing Assets: Transform marketing visuals while preserving their composition, adding new styles and elements with structural guidance.
Adapt User-Generated Content: Repurpose user-generated content for marketing campaigns, making it fit seamlessly with brand aesthetics.
For practical tips, view our guide here. This API endpoint supports content moderation via an optional parameter that can prevent generation if input images contain inappropriate content, and filters out unsafe generated images - the first blocked input image will fail the entire request.
A publicly available URL of the structure reference image. If both structure_image_url and structure_image_file are provided, structure_image_url will be used. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB.
The image file containing the structure reference, in base64 format. This parameter is used if structure_image_url is not provided. Accepted formats are jpeg, jpg, png, webp. Maximum file size 12MB.
The influence of the structure reference on the generated image. This parameter is optional. Higher value means more adherence to the reference structure.
How many images you would like to generate. This parameter is optional. When fast=false, only num_results=1 is supported.
Determines the response mode. When true, responses are synchronous. With false, responses are asynchronous, immediately providing URLs for images that are generated in the background. Use polling for the URLs to retrieve images once ready. This parameter is optional. When fast=false, it is recommended to use sync=false.
Determines the generation mode. When true, the generation will utilize the fast mode which provides the best balance between speed and quality. When false, the regular mode will be utilized. At the moment, tailored models trained using the 'Max' training version, do not support this paramter.
You can choose whether you want your generated result to be random or predictable. You can recreate the same result in the future by using the seed value of a result from the response with the prompt and other parameters. This parameter is optional.
When set to true, generates images with richer details, sharper textures, and enhanced clarity.
Slightly increases generation time per image.
When enabled (default: true), the input prompt is scanned for NSFW or ethically restricted terms before image generation. If the prompt violates Bria's ethical guidelines, the request will be rejected with a 408 error.
When enabled, applies content moderation to both input visuals and generated outputs.
For input images:
- Processing stops at the first image that fails moderation
- Returns a 422 error with details about which parameter failed
For synchronous requests (sync=true):
- If all generated images fail moderation, returns a 422 error
- If some images pass and others fail, returns a 200 response with successful generations and "blocked" objects for failed ones
For asynchronous requests (sync=false):
- Failed images are replaced with zero-byte files at their placeholder URLs
- Successful images are stored at their original placeholder URLs
Flags prompts with potential IP content. If detected, a warning will be included in the response.
The ID of the tailored model to use for generation. This parameter is optional.
The influence of the tailored model on the generation. Only relevant if tailored_model_id is provided. This parameter is optional. When the training data was minimal or less diverse, lower the influence to improve flexibility in the results.
This is relevant only when a tailored model is being used. When true, the model's generation prefix is automatically prepended to your prompt to maintain consistency with the training data, while false allows you to override the training prefix and write the complete prompt yourself, including any preferred prefix text.
https://engine.prod.bria-api.com/v1/reimagine
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
https://engine.prod.bria-api.com/v1/reimagine \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "a book",
"structure_image_url": "URL",
"num_results": 2,
"sync": true,
"fast": true
}'Successful operation.
There are multiple objects in this array (based on the amount specified in num_results) and each object represents a single image or a blocked result.
If you want to recreate the result again, you should use in the request the prompt and the seed of the response.
A list containing a single URL where the generated image can be found. Always includes exactly one URL. When sync=false, the request is asynchronous — the URL is returned immediately, but the image will become available once the inference process is complete.
{ "result": [ { "urls": [ "https://storage.server/generate_image/some_uuid/seed_111111.png" ], "seed": 111111, "uuid": "some_uuid_111111" }, { "urls": [ "https://storage.server/generate_image/some_uuid/seed_222222.png" ], "seed": 222222, "uuid": "some_uuid_222222" } ] }
Request
Description
The /prompt_enhancer route is designed to boost users' creativity by transforming simple prompts into more detailed and vivid descriptions. This helps generate richer, more diverse images. (It is also available as a built-in flag in all of our /text-to-image routes, excluding tailored generation.)
We recommend using this feature by offering users a range of prompts to choose from before generating an image, enabling them to explore creative ideas.
*Works best with short to medium prompts of up to approximately 50 words.
Examples:
original prompt: A cat
enhanced prompt: A black and white photograph of a sophisticated Siamese cat, sitting in a chair next to a large window, with the urban cityscape visible in the background
https://engine.prod.bria-api.com/v1/prompt_enhancer
- curl
- JavaScript
- Node.js
- Python
- Java
- C#
- PHP
- Go
- Ruby
- R
- Payload
curl -i -X POST \
https://engine.prod.bria-api.com/v1/prompt_enhancer \
-H 'Content-Type: application/json' \
-H 'api_token: string' \
-d '{
"prompt": "an old man fishing"
}'{ "results": { "prompt variations": "A close-up, detailed illustration of a fluffy white cat sitting on a windowsill, with its eyes wide open, surrounded by soft, warm light, and a few books scattered nearby" } }