KoboldCpp

A one-file, no-install program for running GGUF models locally, with the KoboldAI Lite web UI for chat, stories and roleplay, plus image and speech models.

12.1k stars, 805 forks C++ On GitHub since March 2023

Visit website View on GitHub How to run it

What is KoboldCpp?

KoboldCpp is built on llama.cpp and inspired by KoboldAI. You download a single executable for Windows, Linux or macOS, pick a GGUF model and get a browser interface for chat, instruct mode, adventure games and collaborative story writing. Beyond text, it can generate images, video, speech and music with compatible models, and it exposes local APIs that front ends like SillyTavern can use. It runs on CPU or GPU.

Good fit if

  • You want to run local models with no installation or dependencies.
  • You write fiction or play text adventures with AI.
  • You want one local server for text, images and speech.

Look elsewhere if

  • You want a plain assistant app with a model library: Jan or Ollama.
  • You need a team service: Open WebUI.

Before you adopt it

KoboldCpp is AGPL-3.0 and entirely free. It is led by its creator, Concedo, who writes about half of the commits, with more than 30 people contributing recently.

Drafted with AI from the project's README and repository data, then checked against that data on 30 October 2026. Report a correction

KoboldCpp health: is it still maintained?

93 of 100, actively developed

  • Activity
    2,260 commits in the last 90 days, the latest on 1 November 2026.
    30 / 30
  • Contributors
    44 people wrote the latest 100 commits; the most active wrote 32% of them.
    20 / 20
  • Releases
    15 releases in the last 12 months; the latest, v1.124.1, on 31 October 2026.
    19 / 20
  • Replies to issues
    Maintainers answered 89% of recent issues, with a median first reply after 4 hours.
    19 / 20
  • Security practices
    No OpenSSF Scorecard for this repository yet, so this part gets a neutral score.
    5 / 10

Read from GitHub on 1 November 2026. How the score works

KoboldCpp license: is it open source?

KoboldCpp uses AGPL-3.0, an open-source license approved by the OSI.

  1. March 2023First published on GitHub.
  2. TodayAGPL-3.0.

How to run KoboldCpp yourself

Docker
Dockerfile or compose file in the repository
Setup effort
1 of 5, judged from the compose file
Kubernetes
No Helm chart in the repository
Desktop and mobile
Builds for Linux, macOS, Windows, including ARM in GitHub releases
Interface languages
English only

Install instructions: koboldcpp.net or the README on GitHub.

Alternatives to KoboldCpp

  • OllamaGet up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models. 89 health · MIT · 186k stars
  • vLLMA high-throughput and memory-efficient inference and serving engine for LLMs 84 health · Apache-2.0 · 96k stars
  • DocsGPTPrivate AI platform for agents, assistants and enterprise search. Built-in Agent Builder, Deep research,… 84 health · MIT · 18.4k stars
  • LocalAILocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware.… 75 health · MIT · 50.1k stars
  • JanJan is an open source alternative to ChatGPT that runs 100% offline on your computer. 73 health · Apache-2.0 · 45.2k stars

All 36 projects in AI chat & local LLMs

Questions about KoboldCpp

Is KoboldCpp open source?

Yes. KoboldCpp is published under AGPL-3.0, an OSI-approved open-source license.

Is KoboldCpp still maintained?

Its health score is 93 of 100 (actively developed). It had 2,260 commits in the last 90 days, the latest on 1 November 2026, and its latest release came out on 31 October 2026.

How do I run KoboldCpp?

You can run it with Docker, as an app for Linux, macOS, Windows, including ARM. The README on GitHub has the exact steps.

What are the alternatives to KoboldCpp?

Close alternatives in the AI chat & local LLMs category include Ollama, vLLM, DocsGPT.