Tabby, my free "Copilot"
summary
Tabby is a free open‑source code assistant you can deploy locally or on your servers via Docker, Homebrew or Bento Cloud and it runs offline after the first model download. It integrates with VS Code, IntelliJ and LSP editors, uses Hugging Face models like StarCoder 1B/3B/7B, and supports repo/doc indexing for context and about 500 tech docs. A single GPU (Nvidia T4) can serve ~50 developers on one process, setup takes 5–10 minutes, admin portal is free for up to five users; expect manual LLM config, security care, and heavy CPU/GPU needs.
AI-written summary · about
chapters (7)
- 0:00Introduction to Tabby
- 3:47What is an AI code assistant
- 7:48Introducing Tabby ML
- 11:33Setting up Tabby
- 14:55Tabby demo and features
- 25:06Key considerations for Tabby
- 28:01Takeaways and conclusion
related talks
24mTabby, my free 'Copilot'
16mTabby, my open Copilot
42mTabby, My Free 'Copilot'
47mFuture-Proof Development with GitHub Codespaces, Copilot, and GPT-4
06h 22mUsing AI to Enhance Developer Productivity
57mAccelerate Your Developer Productivity with AI Embrace the Future Now!
37mEmpowering Developers with local AI
23mFine Tune Your AI Toolbox with GitHub Copilot, Azure AI Foundry...