FICHA · MANJARO

ollama-rocm

Create, run and share large language models (LLMs) with ROCm

  • Service
  • COMMAND-LINE
  • LOCAL-SERVICE
  • AI-MODELS
  • Launchable
  • Runs in terminal
  • Background service
official+codex · reviewed · May 28, 2026 description in en

Description

Provides Ollama with ROCm support for running large language models on compatible AMD GPUs. GPU acceleration can improve local inference speed when hardware and drivers are supported.

Use it when the system has a compatible AMD GPU, ROCm stack, and enough VRAM for the selected models. Unsupported hardware or large models can fail or consume significant resources.

How to run

ollama

Commands: ollama

Permissions

Permissions not analysed for this source yet.