tabby

Tabby provides a self-contained AI coding assistant that operates without the need for external cloud services or a separate DBMS. It is designed to be integrated into existing infrastructure via an OpenAPI interface and supports consumer-grade GPUs, including Apple M1/M2 Metal inference. Key capabilities include: - Code Completion: RAG-based completion utilizing repository-level context and locally relevant snippets. - Answer Engine: A central knowledge engine for engineering teams that integrates with internal data to provide precise answers. - IDE Integration: Extensions available for VSCode, Vim, and JetBrains IDEs, featuring side-panel chat, inline editing, and auto-generated commit messages. - Enterprise Features: Support for LDAP authentication, GitLab/GitHub SSO, team management, and usage analytics reports. - Deployment: Can be deployed via Docker or SkyServe on various cloud environments.