Rubra

Rubra is an open-source tool designed for local development of AI assistants using a large language model (LLM). Conceived with developers in mind, Rubra offers the convenience and intelligence similar to working with OpenAI’s ChatGPT.

It offers an opportunity to build AI-powered applications in a private, cost-effective manner by allowing developers to work locally, thereby bypassing the need for tokens for API calls.

Rubra includes built-in, fully configured open-source LLMs, and it mainstreams the development process for modern AI-powered agents that can interact and process data from multiple channels locally.

It includes a user-friendly chat UI for developers to interact with their models and assistants. Different from other model inferencing engines, it implements an OpenAI compatible Assistants API along with an optimized LLM.

It’s designed with privacy in mind, as it executes all processes on the user’s local machine, ensuring that chat histories and retrieved data never exit the local machine.

Furthermore, Rubra is not restricted to its local LLM; it also supports OpenAI and Anthropic models. Community participation is encouraged, with provisions for user contributions in form of discussions, bug reporting and code contributions on its Github repository.