Skip navigation
llama.vscode (powered by Qwen Coder)
0:00 / 0:37

llama.vscode (powered by Qwen Coder)

Georgi Gerganov

@ggerganov63K followers

Visit on X
468 likesDownloadOriginal post

77,668 views

llama.vscode (powered by Qwen Coder)

9 Comments

Comments from the original post on X (Twitter)

  • Georgi Gerganov's profile picture

    @ggerganov

    This is a lightweight and very efficient VS Code extension using llama.cpp directly to provide local LLM-assisted code and text completions:

    60 likes

  • Georgi Gerganov's profile picture

    @ggerganov

    The llama.cpp server provides unique context reuse techniques that allow you to efficiently use large contexts to enhance the completions based on the contents of your codebase. The setup is simple, no RAG is necessary and the performance is good even on low-end hardware. Enjoy!

    32 likes

  • malico.'s profile picture

    @yondifon

    nvim?? 🥹🥹

    3 likes

  • Georgi Gerganov's profile picture

    @ggerganov

    i got you pal

    4 likes

  • Daniel Nguyen ⚡'s profile picture

    @daniel_nguyenx

    Great work. Thanks

    1 like

  • Neil Chudleigh's profile picture

    @neilsuperduper

    Lets go! Nice work.

  • Nikita 🤙's profile picture

    @Hormold

    Great start!

  • Raymond Weitekamp's profile picture

    @raw_works

    wow - thank you so much for giving the different suggestions for different hardware (RAM). two follow up questions: - can i do this on PC? (my PC happens to have way more GPU/RAM) - from a purely personal perspective - what is the minimum "useful" RAM for this?

  • Georgi Gerganov's profile picture

    @ggerganov

    Yes, this runs on Mac, Linux and Windows - see the setup instructions in the readme. The 7B models are pretty good, so if you can run this (i.e. ~8GB of VRAM) then go for this. Otherwise - use the 3B model (~4GB)

    3 likes