Overview

Own your AI with Git AI.

What is Git AI?

Git AI is a single-interface platform to build and deploy AI. Customers can accelerate their AI deployment, scale and own their AI on any cloud, public or private starting with a library. Git AI is a Git-native portal and command line interface to build, run and track experiments a single configuration.

Why Git AI?

Git is where developers feel at home. Over 95% of developers worldwide use Git and Git is at the core of all AI workflows from development to deployment. Git AI makes the transition for existing AI developers seamless and lowers the barrier to adoption. Git AI offer open APIs offering customers sovereignty over their code, model and data.

Examples of how customers can benefit from Git AI:

  • Git AI is readily available on our partners GPU Clouds and on public clouds like Lambda Labs and Phoenix Systems clouds. Git AI is also readily available for private Kubernetes supporting libraries.
  • Start with an existing library of models that include an open-source LLM, transformers and image synthesizers.
  • Enjoy AI sovereignty by taking your experiments anywhere from existing cloud vendors to on-premises with open APIs.

Licenses

The Git AI library is available for free under permissible licenses Elastic License v2. You can use, extend, or deploy models from the library as long as you do not provide Git AI as a managed service to another party.

Product

Git AI is available with our GPU cloud partners and through our website hosted on public cloud. Please give us feedback at support@codedepot.ai.

List the cloud services:

  • A web portal where users can start and track experiments using the user interface. The web portal also offers tools for users to compare models and experiments.
  • A command line interface where users can create and manage Git AI repositories using a Git-native API.
  • A command line interface where users can create and manage training experiments as well as manage their clusters.
  • A library that users can use to instrument their experiments.
  • A GraphQL API and a runtime for users to query their data available through the web portal and the command line tools.