Local runtime - LostRuins (Concedo) / KoboldAI community - verified
KoboldCpp: install, API, platforms and features
KoboldCpp is a local LLM runtime from LostRuins (Concedo) / KoboldAI community. As its docs describe it: People who want to run GGUF models locally from a single executable with no installation, using a bundled KoboldAI Lite web UI for chat, story writing and roleplay plus a local API for other apps. The latest release is v1.123.1, published October 13, 2026. License: AGPL-3.0 (open source).
- Latest release
- v1.123.1
- Released
- 2026-10-13
- License
- Open source
- OpenAI API
- Yes
Install
Copied verbatim from the page linked under each command. Labels are worded as the source words them.
linux
curl -fLo koboldcpp-linux-x64 https://github.com/LostRuins/koboldcpp/releases/latest/download/koboldcpp-linux-x64 && chmod +x koboldcpp-linux-x64 && ./koboldcpp-linux-x64windows
cmd /c "curl -fLo koboldcpp.exe https://github.com/LostRuins/koboldcpp/releases/latest/download/koboldcpp.exe && koboldcpp.exe"source
git clone https://github.com/LostRuins/koboldcpp.git
cd koboldcppSource: github.com/LostRuins/koboldcpp#compiling-koboldcpp-from-source-code
Run a model
The first-run example from the official docs. OpenAI-compatible endpoint: http://localhost:5001/v1 (default port 5001, changed with --port)
./koboldcpp-linux-x64 --model /path/to/model.ggufFeatures
What the KoboldCpp docs say it supports. Not stated means the docs do not say either way.
Models to run with KoboldCpp
Browse the open-weight models you can run locally for download sizes, licenses and memory fit; check which formats KoboldCpp loads in the table above.
Compare KoboldCpp
FAQ
Does KoboldCpp have an OpenAI-compatible API?
Yes. Endpoint, per its docs: http://localhost:5001/v1 (default port 5001, changed with --port). Source: https://github.com/LostRuins/koboldcpp#apis-and-integrations
Does KoboldCpp have an Anthropic-compatible API?
Which platforms and GPUs does KoboldCpp support?
Its docs list Windows, Linux and macOS, with these backends: CPU, CUDA, Vulkan, Metal and ROCm.
How do I install KoboldCpp?
On linux: "curl -fLo koboldcpp-linux-x64 https://github.com/LostRuins/koboldcpp/releases/latest/download/koboldcpp-linux-x64 && chmod +x koboldcpp-linux-x64 && ./koboldcpp-linux-x64", as shown at https://github.com/LostRuins/koboldcpp#linux. This page lists 3 install commands: linux, windows and source.
Is KoboldCpp open source?
Yes. It is released under AGPL-3.0; the license is at https://github.com/LostRuins/koboldcpp#license.
What is the latest version of KoboldCpp?
v1.123.1, published October 13, 2026 (https://github.com/LostRuins/koboldcpp/releases/tag/v1.123.1), as of October 18, 2026.
Which model formats does KoboldCpp run?
GGUF and Other, per https://github.com/LostRuins/koboldcpp#obtaining-a-gguf-model
Sources
Last verified from the pages below.
- GitHub README - github.com
- Latest release (v1.123.1) - github.com
- KoboldCpp wiki - github.com
- KoboldCpp API reference - lite.koboldai.net
Newsletter
Local AI, every week
One email per week. New model releases, runtime updates and what they mean for what you build.
Keep exploring
Open-weight models you can run
Download sizes, licenses and the run command for each model.
All local LLM runtimes
Every runtime in one feature matrix, with sources.
Picking a runtime for coding agents
Ollama, LM Studio, vLLM and llama.cpp for agent workloads.
Setup guides
Point your coding tools at the model you picked.