You only pay for what you use on Replicate, billed by the second. When you don’t run anything, it scales to zero and you don’t pay a thing.

Hardware Price GPU CPU GPU RAM RAM
- 4x - 8GB
Nvidia A100 (80GB) GPU
1x 10x 80GB 144GB
2x Nvidia A100 (80GB) GPU
2x 20x 160GB 288GB
4x Nvidia A100 (80GB) GPU
4x 40x 320GB 576GB
8x Nvidia A100 (80GB) GPU
8x 80x 640GB 960GB
Nvidia A40 (Large) GPU
1x 10x 48GB 72GB
2x Nvidia A40 (Large) GPU
2x 20x 96GB 144GB
4x Nvidia A40 (Large) GPU
4x 40x 192GB 288GB
8x Nvidia A40 (Large) GPU
8x 48x 384GB 680GB
Nvidia A40 GPU
1x 4x 48GB 16GB
Nvidia T4 GPU
1x 4x 16GB 16GB

If you’re new to Replicate, you can try us out for free, but eventually you’ll need to enter a credit card.

Public models

Thousands of open-source machine learning models have been contributed by our community and more are added every day. When running or training one of these models, you only pay for time it takes to process your request.

Each model runs on different hardware and takes a different amount of time to run. You’ll find estimates for how much they cost under "Run time and cost" on the model’s page. For example, for stability-ai/sdxl:

This model costs approximately $0.012 to run on Replicate, but this varies depending on your inputs.

Predictions run on Nvidia A40 (Large) GPU hardware, which costs $0.000725 per second. Predictions typically complete within 17 seconds.

Image models

Language models

Replicate hosts a selection of language models, including Llama 3 and Mistral, which are priced per token.

Check out our docs for more information about how per-token pricing works on Replicate.

Private models

You aren’t limited to the public models on Replicate: you can deploy your own custom models using Cog, our open-source tool for packaging machine learning models.

We automatically generate an API server for your model and deploy it on a big cluster of GPUs. If you get a ton of traffic, we automatically scale to handle the demand. If you don’t get any traffic, we scale down to zero and don’t charge you a thing.

Unlike public models, you’ll pay for setup and idle time in addition to the time it spends processing your requests.

Learn more

For a deeper dive, check out how billing works on Replicate.