RUNTIME / EASY

KoboldCpp

One-file GGUF chat on Windows, macOS or Linux with no Python install

一手来源

DifficultyEasySetup profile

Platforms5cpu · nvidia · amd · intel · apple

APIOpenAI + Ollama + KoboldCpp-compatible:5001

01 / INSTALL & RUN

安装方式取决于操作系统。以下命令是示例或模板，请先替换占位符并核对量化版本、文件格式、驱动和运行时支持；内存适配不保证部署成功。

1
InstallDownload the single-file release binary from the official GitHub releases (koboldcpp.exe on Windows)
2
Start a modelkoboldcpp --model <downloaded-model.gguf> --usecuda
3
Connect your app
OpenAI + Ollama + KoboldCpp-compatible · :5001. Keep the service bound to localhost unless you add authentication and network controls.

注意

The project warns that koboldcpp.com is an impersonation, so only GitHub release binaries are official. GPU flags differ by vendor and OS, and you can only use the accelerator libraries the release bundles.

02 / COMPATIBLE MODELS