Not out yet.
Training on motion teaches a model how the world holds together. That shows up in stills as light and material behaviour that does not need explaining away.
Campaign key art
The still the whole campaign is built from, in the same look as the video that follows it.
Editorial and illustration
Wide stylistic range, so it is not locked into one house photographic look.
Multilingual creative
Text that renders correctly in more than one script, for markets that are not all English.
Storyboard frames
Generate the frames, then hand the keepers to FLUX 3 Video as keyframes.
Product and lifestyle stills
Grounded lighting and materials, which is most of what makes a product shot believable.
- Availability
- Not released. No date announced by Black Forest Labs
- Available instead
- Flux 2 Klein 9B, Flux 2 Pro, Flux 2 Max
- Provider
- Black Forest Labs
- Family
- FLUX 3, the same model behind FLUX 3 Video
- Training
- Multimodal, on images, video and audio together
- Text rendering
- Accurate, multilingual, inside the image
- Styles
- Wide range, from photographic to illustrative
- Modes
- Text to image and image editing
- Aspect ratios
- Wide range of ratios and resolutions
- Prompts
- Handles complex multi constraint prompts
- Cost
- Not yet priced

FLUX 3 Video
The same model, for motion with sound.

GPT Image 2.5 Flare
Text is the hardest part of the frame and you want the strongest model for it.
What is FLUX 3 Image?
FLUX 3 Image is the announced but unreleased image generation and editing part of Black Forest Labs' FLUX 3 family. It comes from the same multimodal model as FLUX 3 Video, trained on images, video and audio together, and it is expected to generate and edit images across a wide range of styles with accurate multilingual text rendering. Only FLUX 3 Video from that family has actually shipped, so nothing here can run it yet.
How is it different from Flux 2 Max?
FLUX 3 Image is the newer generation and comes from a multimodal model rather than an image only one, which shows in how it handles light, materials and complex prompts. Flux 2 Max remains the established option and is available on Plus and above.
Does it match FLUX 3 Video?
That is the reason to use it. Both come from the same FLUX 3 model, so a still and a clip generated from the same prompt share a look, which matters when both go in the same campaign.
Can it render text?
Yes, accurately and in multiple languages, generated as part of the image. For the most demanding typography, GPT Image 2.5 Flare is still the strongest model available here.
Can I use FLUX 3 Image today?
No. Black Forest Labs has released FLUX 3 Video from this family but not the image model, so it is not selectable on APImage and is not part of any plan. Use Flux 2 Pro or Flux 2 Max for Black Forest Labs stills in the meantime, or GPT Image 2.5 Flare when the frame contains text.
How much will FLUX 3 Image cost?
There is no price yet. Black Forest Labs has not published a rate and the model is not billable on APImage, so any number quoted now would be a guess. It will be priced when it ships, and it will run on the same subscription and the same pool of credits as every other model here.
Do I need a separate subscription?
No. Every model on APImage runs on one subscription and one pool of credits, with no per model surcharge and no extra login.
How do I use it?
FLUX 3 Image is not selectable yet. These are the steps for Flux 2, the Black Forest Labs still model you can run today. Open the image studio and pick Flux 2 Pro, or Flux 2 Max when the hero shot has to hold up at full size. Describe the frame and add references for product, likeness or look. Generate, then send the keeper into FLUX 3 Video as a keyframe if it needs to move.
