Collect Video Testimonials That Build Trust
FLUX 3 Image
FLUX 3 Image by Black Forest Labs generates and edits images from text prompts, supports up to 10 reference images, bounding box composition control, web grounding, and 4K output through an API.
About FLUX 3 Image
FLUX 3 Image is an AI image generation and editing model developed by Black Forest Labs. Released on October 1, 2026, it enables users to generate images from text descriptions, modify existing images, and create new compositions using multiple reference images. The model is part of the FLUX 3 family, which is built on a multimodal foundation architecture.
The model supports text to image generation, instruction based image editing, and compositions using up to 10 reference images. Users can specify subjects, settings, lighting, and framing through natural language prompts. Its bounding box functionality allows individual elements to be positioned within an image or targeted for local modifications, including recoloring, replacement, movement, resizing, and removal.
FLUX 3 Image supports five output resolution settings, from 768 square pixels to approximately 16 megapixels at 4K resolution. It provides 15 predefined aspect ratios and an automatic aspect ratio option. Reference images can be supplied through URLs or base64 encoded data, with each image supporting resolutions between 256 × 256 pixels and 16 megapixels.
The model is accessible through the Black Forest Labs API and a browser based playground. Its API supports configurable safety tolerance and optional web and image search grounding. FLUX 3 Image is intended for developers, designers, creative professionals, and businesses building image generation, editing, and visual content workflows.
Key Features
-
Text to Image Generation: Generates images from natural language prompts describing subjects, environments, lighting, styles, and composition.
-
Instruction Based Image Editing: Modifies existing images using text instructions, supporting targeted changes to objects and visual elements.
-
Multiple Reference Images: Accepts up to 10 reference images for image editing, style changes, and combining visual elements into new compositions.
-
Bounding Box Composition Control: Supports positioning objects and text elements using bounding box coordinates defined on a 0 to 1000 grid. Bounding boxes also enable targeted edits to selected image regions.
-
4K Image Generation: Supports 768sq, 1k, 1.5k, 2k, and 4k resolution settings, with output reaching approximately 16 megapixels. Provides 15 predefined aspect ratios and an automatic option.
-
API Access and Search Grounding: Provides a single API endpoint for generation and editing, with optional web and image search grounding and configurable safety tolerance levels from 0 to 4.
Pricing
FLUX 3 Image uses API pricing based on the resolution of each generated image.
- 768sq Image Generation: $0.041/image
Image generation at the 768sq resolution setting.
- 1K Image Generation: $0.048/image
Image generation at the 1k resolution setting.
- 2K Image Generation: $0.100/image
Image generation at the 2k resolution setting.
- 4K Image Generation: $0.607/image
Image generation at the 4k resolution setting, supporting approximately 16 megapixels.
- 1.5K Image Generation:
The 1.5k resolution setting is supported, but its individual price is not specified in the official release notes.
Pricing last updated: October 12, 2026 at 12:00 AM
Use Cases
-
Generate marketing graphics, illustrations, posters, and other visual content from text descriptions.
-
Edit existing images by recoloring, replacing, moving, resizing, or removing selected elements.
-
Combine reference images to create compositions incorporating specified objects, scenes, and styles.
-
Build image generation and editing applications with configurable layouts, resolution settings, and reference images through the API.
Pros & Cons
Pros:
-
Supports image generation, editing, and up to 10 reference images through a single API endpoint.
-
Provides bounding box controls for element placement and localized image modifications.
-
Supports output resolutions up to 4K, 15 predefined aspect ratios, and optional web and image search grounding.
Cons:
-
Accepts a maximum of 10 reference images per request, with each reference limited to 16 megapixels.
-
Generated image results must be retrieved through the polling workflow, and the documented result download is available for one hour.
Integrations
Black Forest Labs API, FLUX 3 Image REST API, Black Forest Labs Playground, Python, JavaScript, cURL
FAQ
What is FLUX 3 Image?
FLUX 3 Image is an AI image generation and editing model developed by Black Forest Labs. It generates images from text prompts, edits existing images, and combines up to 10 reference images into new compositions.
When was FLUX 3 Image released?
FLUX 3 Image was released on October 1, 2026, according to the official Black Forest Labs release notes. The broader FLUX 3 model family was announced in July 2026.
Does FLUX 3 Image support text to image generation?
Yes. FLUX 3 Image generates images from natural language prompts describing subjects, settings, lighting, framing, and other visual details.
Can FLUX 3 Image edit existing images?
Yes. The model supports instruction based image editing. Users can specify changes to existing images and use bounding boxes to target individual elements for recoloring, replacement, movement, resizing, or removal.
How many reference images can FLUX 3 Image use?
FLUX 3 Image supports up to 10 reference images in a request. Images can be provided through URLs or base64 encoded data, with each reference supporting resolutions from 256 × 256 pixels to 16 megapixels.
What is the maximum image resolution supported by FLUX 3 Image?
FLUX 3 Image supports output resolution settings of 768sq, 1k, 1.5k, 2k, and 4k. Its maximum documented output is approximately 16 megapixels at the 4k setting.
What are bounding boxes in FLUX 3 Image?
Bounding boxes define the locations of image elements using top, left, bottom, and right coordinates on a 0 to 1000 grid. They allow users to control composition and specify regions for localized image edits.
Does FLUX 3 Image support web search grounding?
Yes. FLUX 3 Image supports optional web and image search grounding before generation. Grounding is enabled by default through the API and can be disabled using the grounding parameter.
How can developers access FLUX 3 Image?
Developers can access the model through the Black Forest Labs REST API using the POST /v1/flux-3-image endpoint. Requests require an API key, and the same endpoint supports generation and editing. Black Forest Labs also provides a browser based playground.
Does FLUX 3 Image generate video or audio?
FLUX 3 Image is documented for image generation and editing. Although the broader FLUX 3 family includes video and audio capabilities, the FLUX 3 Image endpoint is specifically documented for image tasks.
Last edited
October 12, 2026 at 9:20 AM by Venkatraman Chandrasekaran
Similar to FLUX 3 Image
