FICHA · AUR

llama.cpp-clblast

Port of Facebook's LLaMA model in C/C++(with Unofficial CLBlast Backend support)

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

Description

OpenCL-class GPU acceleration through CLBlast can speed up local LLM inference on supported hardware.

This package is useful for users who specifically need the unofficial CLBlast backend for llama.cpp. It is an inference build variant, not a separate model or chat application.

GPU backends can be sensitive to driver versions and may use significant power or memory. Test stability before long runs.

How to run

llama-cli

Commands: llama-cli

Permissions

Permissions not analysed for this source yet.