About fal
fal is a platform for generative media models. It offers one API for image, video, audio and 3D models, along with serverless GPUs for deploying your own.
What it does
Developers call hosted models for tasks such as text to image, image editing, video generation and speech. The platform also lets teams deploy and train their own models on serverless GPU infrastructure. New models are added as they launch, so you can test the latest options without setting up separate providers.
Key features
- Catalog of over 1,500 generative media models
- Text to image, image editing and video models
- Audio, voice and 3D model access
- Serverless GPUs for custom models
- Training and deployment for production use
Who it's for
- Developers adding AI media features to products
- Creative software teams
- Studios that need several media models in one place
Pricing and good to know
Models are billed per use, and rates depend on the model. The site lists a pricing page and an enterprise contact option for larger accounts. Compare the cost per output for your use case before you scale, and check output quality on your own prompts, since model results vary.
Highlights
- Call over 1,500 generative media models
- Generate and edit images
- Create video and audio from prompts
- Run serverless GPUs for custom models
- Train and deploy models for production
