This KoboldCpp variant runs GGML models with AMD ROCm offloading and the KoboldAI interface.
KoboldCpp
KoboldCpp runs GGUF models through a KoboldAI interface in a single-file application.

About KoboldCpp
KoboldCpp is free, open-source software for running GGUF large language models on your own computer. Users can chat with an AI assistant, write stories, do roleplay, or connect other apps to a local API. It runs on CPU or GPU and is distributed as a single executable that requires no installation. Ready-to-run downloads are listed for Windows, Linux, and macOS.
The software also includes text, image, video, speech, and music generation with compatible models, an integrated agent, and a bundled KoboldAI Lite WebUI. Which generation types are available depends on the models used, and the integrated agent must be started manually through the launcher or a command-line flag.
Key features
- Runs GGUF language models for chat, adventure, instruct, and story writing modes
- Image generation and editing with Stable Diffusion, SDXL, Flux, and other models
- Speech-to-text with Whisper and text-to-speech with models such as Kokoro and Dia
- Tool calling, MCP server support, web search, and retrieval-augmented generation through TextDB
- Integrated KoboldCpp Agent for writing and editing code, running programs, and scheduling tasks
- Compatible API endpoints for KoboldAI, OpenAI, Ollama, A1111/Forge, and ComfyUI
- Free and open-source software
- Distributed as a single executable that requires no installation
- Runs on CPU or GPU
- Video, speech, music, and image generation depend on compatible models
- The integrated agent must be started manually from the launcher's Admin tab or with the --agent flag
KoboldCpp at a glance
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
KoboldCpp alternatives
Tools and sites you could use instead, ranked by how closely they match.
Ollama helps users run and work with open models through a local AI tool.
SillyTavern is a frontend for users working with large language models.
llamafile packages language models so they can be distributed and run as a single file.
This HR chatbot answers user questions with tools through an autonomous agent workflow.
This project describes serving an LLM as a chatbot service through a multi-agent framework.
Popular in AI Chatbots & Assistants
Well-known listings from the same topic.
KoboldCpp FAQ
How do I start the integrated KoboldCpp Agent?
Does KoboldCpp need to be installed?
Which operating systems have ready-to-run downloads?
Tell us and an editor will check it.
Claim the listing and suggest updates yourself. An editor checks every claim first.












