Sagify

Sagify provides a simplified interface to manage machine learning workflows on AWS SageMaker, helping you focus on building ML models rather than infrastructure. Its modular architecture includes an LLM Gateway module to provide a unified interface for leveraging both open source and proprietary large language models. The LLM Gateway gives access to various LLMs through a simple API, letting you easily incorporate them into your workflows.

Sagify is designed to streamline ML development by handling cloud infrastructure tasks. It focuses on moving from idea to deployed model quickly while reducing the engineering overhead associated with training, tuning, and deployment. The tool emphasizes enabling teams to concentrate on model development and experimentation, rather than managing complex deployment pipelines.

Sagify offers a set of capabilities to work with both proprietary and open-source LLMs, and to deploy them via AWS SageMaker endpoints. It supports role-based workflows for exploration and production use, with a unified API to access different providers and models.

Sagify documents prerequisites and deployment options, including local Docker-based Gateway and AWS Sagemaker-backed deployments. It describes how to configure AWS and run deployment commands, and outlines steps for starting the LLM Gateway server and deploying models across platforms.

Use cases

• Experimenting with LLMs (proprietary or open-source) and deploying them on AWS SageMaker.
• Managing ML workflows from data preprocessing to model deployment in the cloud.