KoboldCpp
Single-file llama.cpp-based runner for GGUF models with a story-writing UI.
- Stars
- 12K
- Forks
- 791
- Open issues
- 528
- Last commit
- 2026-10-15
Live from the GitHub API · updated 2026-10-15
What is KoboldCpp?
KoboldCpp packages llama.cpp into one executable with the KoboldAI Lite UI for chat and storytelling, plus image generation, speech and an API compatible with OpenAI and KoboldAI clients.
Key features
- One executable, no install
- Story, adventure and chat modes
- Bundled image generation and TTS
- Works as a backend for SillyTavern
System requirements
- Platforms
- Windows · macOS · Linux
- GPU
- Optional (CUDA, Vulkan, Metal)
- RAM
- Depends on model size
How to install KoboldCpp
Summarized from the project’s README. Projects change quickly — check the GitHub page for the latest instructions.
Windows
- 1Download koboldcpp.exe from GitHub Releases and run it.
- 2Select a GGUF model in the launcher and click Launch.
Linux
- 1Download the binary and run it.
curl -fLo koboldcpp https://github.com/LostRuins/koboldcpp/releases/latest/download/koboldcpp-linux-x64 && chmod +x koboldcpp && ./koboldcpp
Strengths
- Zero-setup
- Great for roleplay and fiction
Limitations
- UI is niche-focused
KoboldCpp alternatives compared
| Software | Stars | License | Platforms | VRAM |
|---|---|---|---|---|
| KoboldCpp Single-file llama.cpp-based runner for GGUF models with a story-writing UI. | 12K | AGPL-3.0 | — | |
| llama.cpp LLM inference in C/C++ — the engine behind most local AI apps. | 131.2K | MIT | — | |
| Text Generation WebUI The “AUTOMATIC1111 of text generation” — a Gradio UI for local LLMs. | 47.7K | AGPL-3.0 | — | |
| llamafile Distribute and run an LLM as a single executable file. | 26.3K | Apache-2.0 | — | |
| Ollama Get up and running with large language models locally. | 183.2K | MIT | — | |
| Jan Open-source ChatGPT-style desktop app that runs 100% offline. | 45K | AGPL-3.0 | — |
FAQ
Is KoboldCpp free?
Yes — KoboldCpp is open-source software released under the AGPL-3.0 license. Check the license (and any model weight licenses) before commercial use.
Does KoboldCpp run on Windows, Mac and Linux?
KoboldCpp supports: Windows, macOS, Linux.
What hardware do I need for KoboldCpp?
Optional (CUDA, Vulkan, Metal). RAM: Depends on model size
How do I install KoboldCpp?
1. Download koboldcpp.exe from GitHub Releases and run it. 2. Select a GGUF model in the launcher and click Launch.
What are the best KoboldCpp alternatives?
Popular alternatives include llama.cpp, Text Generation WebUI, llamafile, Ollama. See the comparison table on this page.