AI Image Generator
PRODUCT OVERVIEW
FLUX is a family of multimodal generative models from Black Forest Labs. FLUX 3 unifies image, video, audio, and action prediction: it generates stylistically diverse images with accurate text rendering, produces video with optional multilingual speech, effects, and ambience in clips up to 20 seconds from text, images, or keyframes, and supports robotics through visual action prediction. Users can experiment in a browser playground, integrate models through a production-scale API, or run open weights on their own infrastructure with licensing for fine-tuning and customization. Enterprise options add scale, customization, and compliance certifications. Access is pay-as-you-go per generation, with no subscriptions or seat fees.
Open the FLUX dashboard playground to test prompts and transform media in the browser without writing code.
Follow the API documentation to access FLUX models programmatically and handle production workloads at scale.
Download FLUX models from Hugging Face or GitHub, obtain a license, and run, fine-tune, or customize them on your own infrastructure.
Generating stylized images from text or reference images
Producing short video clips with native audio from text, images, or keyframes
Experimenting with prompts in a no-code playground
Integrating generative models into products via API
Self-hosting and fine-tuning models on proprietary infrastructure
Researching robotic action prediction
FLUX uses pay-as-you-go pricing with no subscriptions or seat fees; you pay per image generated based on the resolution tier selected in the request.
FLUX 3 Image supports 768×768, 1K, 1.5K, 2K, and 4K tiers, with rates from $0.041/img to $0.607/img respectively.
No. Reference images and prompts are included at no extra charge, and the per-image rate does not vary with the number of reference images up to the documented maximum.
Rejected requests, failed generations, and outputs withheld by moderation are not charged.
Yes. Open weights are available for self-hosted deployment with licensing options covering fine-tuning, LoRA rights, and customization.
FLUX 3 is a multimodal model covering image, video with optional audio, and action prediction for robotics.
FLUX 3 video generation supports clips up to 20 seconds in a single generation, starting from text, an image, or keyframes.
Yes. Enterprise offers custom pricing, volume discounts, SLA guarantees, dedicated support, and compliance certifications such as SOC 2 and ISO 27001.
Pay-as-you-go pricing per generation; no subscriptions or seat fees. Enterprise custom pricing and open-weights licensing are available on request.
Pay-as-you-go per-image pricing for FLUX 3 Image, charged by the resolution tier selected in the request.
Lower-cost per-image tier for the open-weights FLUX 3 Image Klein model, billed by resolution tier.
Self-hosted license for developers and early-stage teams deploying FLUX models on their own infrastructure.
Self-hosted license for product teams shipping at volume with additional model access.
Self-hosted license for agencies and service providers needing multiple domains.
Custom enterprise plan for larger organizations adding FLUX API and open weights to their products, with custom pricing and volume discounts.
COMMUNITY REVIEWS
No reviews yet.
Unified API gateway for leading image, video, audio, and text generation models, with persistent artifact storage and callbacks.
Run and fine-tune open-source and proprietary AI models via API, with pay-per-use pricing and custom model deployment.
