OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Coding Tools

Llama.vscode

Llama.vscode is an MIT-licensed Visual Studio Code extension that connects editor completion, AI chat and agentic coding workflows to language models, including local models served through llama.cpp.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Llama.vscode — OpenViST project profile in AI Coding Tools.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want model-assisted editing or an agent interface inside VS Code and are prepared to select compatible models and configure a local llama.cpp runtime or another documented model option.

Who it is for

  • VS Code developers who want inline text and code suggestions from language models.
  • Users seeking a local-runtime workflow for completion and chat.
  • Developers who want an editor agent that can receive queries, use attached files and access configured MCP tools.

Documented uses

  • Generate code or text suggestions while editing and accept them incrementally with keyboard controls.
  • Adjust generation-time limits and the context considered around the cursor.
  • Ask a coding agent questions or provide tasks with selected files attached.
  • Choose model groups for completion, chat, embeddings and tools.
Before you choose

Requirements & limits

The repository identifies the project as MIT licensed. The README lists installation through the VS Code Marketplace and Open VSX.

  • Completion requires a fill-in-the-middle-compatible model; the README links to a model collection.
  • Suggested local model configurations vary by available VRAM, and CPU-only examples carry an explicit quality trade-off.
  • Agent workflows can use internal tools and tools from MCP servers installed and started in VS Code.
  • The extension exposes controls for suggestion scope and maximum generation time.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up llama.cpp through Llama.vscode

Install or upgrade the documented local runtime and choose an environment in the extension.

Before you start

  • Llama.vscode installed in VS Code.
  • macOS: Homebrew; Windows: Windows Package Manager (winget).
  • Linux: llama.cpp binaries with the bin folder added to PATH.

Steps

  1. Start Llama.vscode; its README says llama.cpp setup is automatic on startup for supported installation routes.

  2. If needed on macOS or Windows, open the extension menu from the status bar or press Ctrl+Shift+M and choose “Install/Upgrade llama.cpp”.

  3. On Linux, obtain the latest llama.cpp binaries and add their bin folder to PATH.

  4. Open the extension menu and select “Select/start env...” for the intended use.

Expected outcome

The local runtime is installed or available and an extension environment is selected.

Guide limitations

  • The stated automatic installation route lists Homebrew for macOS and winget for Windows; Linux setup requires handling binaries and PATH.
  • Completion requires a fill-in-the-middle-compatible model.
  • The README's CPU-only configurations are described as producing significantly lower quality.
Documentation 1 ↗
Source-based practical guide

Start a Llama Agent session

Open the agent interface, select an agent environment and submit a coding query.

Before you start

  • Llama.vscode installed in VS Code.
  • An environment containing an agent selected or available to select.

Steps

  1. Open Llama Agent with Ctrl+Shift+A or choose “Show Llama Agent” from the extension menu.

  2. Select an environment with an agent if one has not already been selected.

  3. Write a query and, if needed, attach files using the @ button.

Expected outcome

A query is submitted in the Llama Agent interface, optionally with attached files.

Guide limitations

  • The README does not provide a guaranteed result or completion time for agent tasks.
  • MCP tools depend on MCP servers installed and started in VS Code.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for AI coding assistance

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
Llama.vscodeGenerate code or text suggestions while editing and accept them incrementally with keyboard controls.DesktopThe repository identifies the project as MIT licensed. The README lists installation through the VS Code Marketplace and Open VSX.
CodedroidaiGenerate procedures, unit tests, FMX forms and VCL forms.Desktop, Linux, macOS, WindowsThe client connects to OpenAI and Replicate using REST integrations, and the README says API keys are required for that functionality. It describes Windows, macOS and Linux support but also says the client is currently specifically set up for Windows. Replicate models may be run in the cloud or locally through Docker.
Codinit DevGenerate and manage full-stack Node.js application code.Docker, Linux, macOS, WebThe repository documents prebuilt desktop releases for macOS, Windows and Linux, and browser and Electron use. The project is open source under the MIT license.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
ggml-org/llama.vscode · Official homepage title · retained in its source language
How we review sourcesSuggest a correction