Comment on page

Introduction

The Shale-Serve API is a production-ready* inference API for open LLMs (large language models). It’s hosted on a highly scalable GPU cloud infrastructure that the Shale Protocol team has built in-house.
The API is "production-ready,” as defined in MLOps. The Shale Protocol team discloses the underlying models that the API supports. It’s the developers’ responsibility to assess the license of the models and the possibility of commercial use.

Developers only need to obtain an API key to get free access to the Shale-Serve API. No credit card is required. The free tier allows for a maximum of 10,000 daily requests per API key.

Last modified 7d ago