FICHA · AUR

llama.cpp-hip

Port of Facebook's LLaMA model in C/C++ (with AMD ROCm optimizations)

  • llm-inference-tool
  • CLI
  • Dev
  • Launchable
  • Runs in terminal
official+codex · reviewed · Jun 2, 2026 description in en

Description

AMD ROCm acceleration can improve local LLM inference on supported GPUs.

This package is useful for users with compatible AMD hardware who want a llama.cpp build tuned for ROCm. It is an inference backend variant, not a model download.

ROCm support depends on GPU, kernel, and driver compatibility. Check the stack before using it for long sessions.

How to run

llama-cli

Commands: llama-cli

Permissions

Permissions not analysed for this source yet.