Give us an upvote on Product Hunt! Here, you will also find a Promo Code to receive 67% off the first month of your Pro Subscription!
Pricing
Choose a plan and start creating workflows
Serverless Pricing
Serverless inference pricing is billed on a per-second basis. Every model’s Pricing Page shows the per-second price for using the model. The run time will depend on a number of factors like number of steps, image size, parameters selected, etc. Click here to buy credits to use for Serverless APIs.
Sample Pricing
Image Models
faceswap-v2
$0.0015/second
Codeformer
$0.0015/second
Stable Diffusion 3 Medium Text to Image
$0.006/second
Video Models
sam-v2-video
$0.0015/second
Video-faceswap
$0.0038/second
LLMs
Llama-v3-405b
$0.0015/second
$0.0015/second
GPR-4-turbo
$0.0015/second
$0.0015/second
Dedicated API Pricing
Choose a Model or Workflow: Select from our options or upload your own.
Configure Machine Availability:
a. Hot Machines: Always on, 24/7.
b. Cold Machines (on standby): Automatically spin up as needed, with a start time of 1-5 seconds, and remain active based on your idle time settings.Customize Settings:
a. Set the idle time threshold.
b. Define the API request timeout.
Sample Pricing
A100 GPUs
Standby Machines (Monthly Fixed Cost)
$0.10/hour
Standby Machines (Usage Cost)
$3.36/hour
Active Machines (Monthly Fixed Cost)
$2.28/hour (Unlimited Usage)
Fine Tuning Pricing
Our fine-tuning service is priced competitively at $0.0020 per second on an A40 GPU. We recommend using ~20 images for fine-tuning. The approximate time to complete fine tuning is around 1-hour and the cost typically ranges from $5-10 per session. Actual cost and time will vary depending on the number of images used to fine-tune and the resolution of those images. To learn more about Fine Tuning, take a look at our Docs Page.