Fast, reliable, and cost-efficient.
fal.ai adapts to your usage, ensuring you only pay for the computing power you consume. It's cost-effective scalability at its best.
For private serverless model pricing, please see our enterprise pricing page.
Pricing
Pricing
API Billing Based on Compute Seconds
Competitive pricing for custom deployments. Get H100s from from as low as $1.99/hr. Contact support, [email protected], to get started.
GPUH100
VRAM80GB
Price per hour*$1.99/h
Price per second*$0.0006/s
GPUH200
VRAM141GB
Price per hour*$2.10/h
Price per second*$0.0006/s
GPUA100
VRAM40GB
Price per hour*$0.99/h
Price per second*$0.0003/s
GPUA6000
VRAM48GB
Price per hour*$0.60/h
Price per second*$0.0002/s
GPUGPU-B200
VRAM192GB
Price per hour*contact us
Price per second*contact us
*starting at
API Billing Based on Model Output
Some models are billed by model output, instead of compute seconds. For more details, see each model page.
| Model Name | Unit Price (USD) | |
|---|---|---|
| Veo 2 Image to Video | 0.25 per second | |
| Hunyuan Video | 0.4 per video | |
| Kling 1.6 Pro Video | 0.095 per second of video | |
| Kling 2 Master Video | 0.28 per second of video | |
| Alibaba Wan Video | 0.4 per video | |
| FLUX.1 [dev] | 0.025 per megapixel | |
| FLUX.1 [schnell] | 0.003 per megapixel | |
| FLUX.1 [pro] | 0.05 per megapixel | |
| Stable Diffusion 3 - Medium | 0.035 per image | |
| Stable Video | 0.075 per video | |
| MiniMax Video Live | 0.5 per video |