Text or image to video
$0.06–$0.80per second of video, pay-as-you-go$0.06/s Draft HD, $0.17/s HD, $0.29/s FHD, $0.40/s QHD, $0.80/s UHD (4K). Native audio included. 5 s of standard HD costs $0.85. Official source 1 ↗
Reviewed tool profile
FLUX 3
Black Forest Labs' multimodal model for video with synchronized audio, image generation and robot actions.
Sources checked
- Pricing
- FLUX 3 is pay-as-you-go with no subscription, billed in credits worth $0.01 each, at the same price in the API and Playground. Text or image to video costs $0.17 to $0.80 a second from HD to 4K ($0.06 in draft); continuation costs $0.41 to $0.95 ($0.12 in draft). FLUX 3 Image costs $0.041 to $0.607 per image. Enterprise pricing is by quote. Official source 1 ↗ · Official source 2 ↗
- Available as
- FLUX 3 runs in the browser through the BFL dashboard (Playground) and is available to developers through the BFL API. Official source 1 ↗ · Official source 2 ↗
- Primary group
- Video & Animation
What it does
FLUX 3 is Black Forest Labs' multimodal foundation model. One model generates video with synchronized audio and generates and edits images, and an open-weight 7B variant, FLUX 3 Action, predicts robot actions. Video and image run through the BFL dashboard (Playground) and API.
Official source 1 ↗ · Official source 2 ↗Features
- Text, image and video to video. Generate clips from a text prompt, animate a still image as the first frame, or continue an existing clip from its last frame. Official source 1 ↗ · Official source 2 ↗
- Native audio and dialogue. Video can come with generated audio in the same pass, including multilingual dialogue with lip sync, sound effects and ambience, at no extra charge. Official source 1 ↗
- 20-second clips up to 4K. Text and image to video run 5–20 seconds and continuation 5–15 seconds per generation, at 24 fps from HD up to 4K, in aspect ratios from 21:9 to 9:21. Official source 1 ↗
- Keyframes. Image to video takes 1–10 images: pin the first frame, a start and end frame, or several ordered keyframes, and FLUX 3 interpolates one continuous shot. Official source 1 ↗ · Official source 2 ↗
- Draft mode. Draft renders a fast, cheaper HD preview; when the direction is right, the same video can be re-rendered at full quality, billed at the standard FHD rate. Official source 1 ↗
- Multiple scenes and agentic chaining. One video can hold several scenes and camera angles, and clips can be chained into longer sequences, for example by an agent that generates, checks and continues each shot. Official source 1 ↗ · Official source 2 ↗
- Video editing with a prompt. FLUX Video Edit, a separate FLUX Tool, changes an existing clip from a text instruction while leaving the rest of the shot as filmed. Official source 1 ↗ · Official source 2 ↗
- Typography and wide style range. Renders on-screen text and handles styles from camcorder footage to animation, not only cinematic looks. Official source 1 ↗ · Official source 2 ↗
- FLUX 3 Image and FLUX 3 Action. The same model family includes FLUX 3 Image, which generates and edits images up to 4K with up to ten references and bounding-box placement, and FLUX 3 Action, a 7B open-weight model that predicts robot actions and a short future video. Official source 1 ↗
Pricing & free limits
Video to video
$0.12–$0.95per second of video, pay-as-you-go$0.12/s Draft HD, $0.41/s HD, $0.53/s FHD, $0.65/s QHD, $0.95/s UHD (4K). Official source 1 ↗
FLUX 3 Image
$0.041–$0.607per image, pay-as-you-go$0.041 at 768 × 768, $0.048 at about 1 MP, $0.100 at about 4 MP, $0.607 at about 16 MP (4K). Official source 1 ↗
Enterprise
Custom quotecontractVolume discounts, custom pricing, SLA guarantees and dedicated support. Official source 1 ↗
Billing details
There are no subscriptions or seat fees; you pay per generated second. Official source 1 ↗
Enhancing a draft into a full-quality result is charged at the standard FHD rate. Official source 1 ↗
Usage is billed in prepaid credits; one credit is $0.01, and API and Playground prices are the same. Official source 1 ↗
Drafts cost about a third of a full render; FHD, QHD and UHD are the same generation upsampled and billed at the delivered class. Official source 1 ↗
Free access & limits
Free access and its limits were not established by the checked official sources.
Best for & limitations
Advertising, product and brand video: BFL lists advertising, virtual try-on and storyboarding among its use cases, and partners use it for on-brand product and motion-design work. Official source 1 ↗
Filmmaking and longer stories: chain 20-second clips with keyframes and continuation into multi-shot sequences that keep characters consistent. Official source 1 ↗
Explainers and documentary-style footage, including dialogue in several languages. Official source 1 ↗ · Official source 2 ↗
Developers building video features into their own products through the BFL API. Official source 1 ↗
- Preview model. BFL labels FLUX 3 a preview model; Omni Reference with images and videos is not available yet. Official source 1 ↗
- Draft mode is HD only. Draft previews cannot be rendered above HD. Official source 1 ↗
- Clip length caps. One generation is capped at 20 seconds (15 seconds for continuation); longer pieces need continuation or chaining. Official source 1 ↗ · Official source 2 ↗
- Prompts may be used for training. Prompts, uploads and outputs can be used to train BFL's models unless you email a training opt-out; enterprise-contract data is excluded. Official source 1 ↗
- Open-weight FLUX 3 Dev not yet released. The open-weight FLUX 3 Dev multimodal backbone for video and image is announced but not released. Official source 1 ↗
Similar tools
Tools like FLUX 3
Published tools that share its categories.
VirseCreate and edit commercial visual work on a canvas with multimodal AI models and aesthetic memory.
Visit 
SparkVidCreate and revise videos from text or images with multiple AI video models in one online workspace.
Visit 
PixfyCreate an AI character from text or a reference image and reuse it across outfits, poses, scenes, products, and brand mascots.
Visit 
InvideoAI-assisted video editing with natural-language directions, a multitrack timeline and real-time collaboration.
Visit 
StivioTurn a still image and a motion prompt into a short HD video using a choice of AI video models.
Visit 
MioraBuild brand-consistent campaign assets with a creative agent, editable canvas, and memory.
Visit