San Francisco, CAWhat's good about fal
- Focus Generative media
- Also sells Raw GPU hours
- Custom code Supported
- Known for FLUX image models
Available GPUs
No GPU pricing data available yet. Check back soon.
About fal
fal runs generative media models, mainly image and video, as hosted APIs billed on output rather than on compute: video is charged per second or per video and images by count or by megapixels of output. Alongside that it sells serverless GPU capacity directly, with a published hourly rate card covering B300 (288GB), B200 (180GB), H200 (141GB), H100 (80GB) and RTX PRO 6000 (96GB), and supports deploying custom applications onto that fleet rather than restricting customers to hosted models. The company is closely associated with the FLUX family of image models.
Pricing options
Output-based (per imageper megapixelper second of video)hourly GPU instances
Services
Generative media model APIsserverless GPU computecustom app deployment
- Generative media Focus
- Image and video models served as APIs, billed on output rather than compute time.
- Raw GPU hours Also sells
- Public hourly rate card covering B300, B200, H200, H100 and RTX PRO 6000.
- Supported Custom code
- Customers can deploy their own applications onto the GPU fleet.
Compliance and certifications
Independently verified standards governing security, privacy and operations.
Ready to deploy on fal?
Visit fal