---
title: "Discussion on \"Running llama.cpp (compiled from source) on AMD Strix Halo 395\" | Hashnode"
description: "Discussion on \"Running llama.cpp (compiled from source) on AMD Strix Halo 395\". Just a quick doc/note/tutorial for referencing myself later. \nHere's how to get llama.cpp running with Vulkan support on AMD AI Max 395 (Strix Halo) based devices. I tried it on a Beelink GTR 9 Pro, but should work for Framework Ddesktop too. \nInstal..."
url: "https://hashnode.com/posts/running-llamacpp-compiled-from-source-on-amd-strix-halo-395/697d6569c1b9799b3479aeb7"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Running llama.cpp (compiled from source) on AMD Strix Halo 395](https://arnav.tech/running-llamacpp-compiled-from-source-on-amd-strix-halo-395?utm_source=hashnode&utm_medium=feed)

Just a quick doc/note/tutorial for referencing myself later. Here's how to get llama.cpp running with Vulkan support on AMD AI Max 395 (Strix Halo) based devices. I tried it on a Beelink GTR 9 Pro, but should work for Framework Ddesktop too. Instal...

Jan 31 · 1 min read · arnav.tech  
Read the article

-   [#vulkan](https://hashnode.com/tag/vulkan)
-   [#llama](https://hashnode.com/tag/llama)
-   [#graphics](https://hashnode.com/tag/graphics)
-   [#drivers](https://hashnode.com/tag/drivers)
-   [#inference](https://hashnode.com/tag/inference)

## Written by

[Arnav Gupta](https://hashnode.com/@arnavgupta)

@arnavgupta

  
[View profile](https://hashnode.com/@arnavgupta)

Android and NodeJS Developer and Teacher. Have founded a few companies.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.