What it does
fal is a generative media platform for developers. Its model gallery lists more than 1,000 image, video, audio and 3D models that you call through one API and SDKs, with no fine-tuning or setup needed. The site also says it gives early access to some new models. For your own models, fal serverless runs inference on GPUs that scale from zero to thousands without you configuring GPUs or an autoscaler. You can deploy private or fine-tuned models, bring your own weights, and monitor workloads. For larger jobs, fal compute offers on-demand VMs and dedicated clusters on NVIDIA H100, H200 and B200 hardware for fine-tuning, training or running custom models. The site lists enterprise features including SOC 2 compliance, single sign-on, private endpoints, usage analytics and 24/7 priority support, and says billing is per output for serverless or per hour for GPUs on compute.
Capabilities
- Gallery of 1,000+ image, video, audio and 3D models behind one API and SDKs
- Serverless GPU inference that scales from zero without autoscaler setup
- Deploy private or fine-tuned models, or bring your own weights
- On-demand H100, H200 and B200 GPU VMs and dedicated training clusters
- Private endpoints, SSO, usage analytics and SOC 2 compliance for enterprise teams
When to use it
- Adding image, video or audio generation to a product through hosted model APIs
- Serving your own or a fine-tuned model without managing GPUs
- Renting H100, H200 or B200 capacity for training or large inference jobs
