julian/amd-strix-halo-toolboxes
1
0
Fork 0
Code Issues Pull Requests 2 Actions Packages Projects Releases Wiki Activity

moved update notice up

Browse Source
This commit is contained in:
Donato Capitella
2025-09-28 09:39:14 +01:00
parent ba88675b9c
commit c2624c5cbe
1 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+6 -3
View File
@@ -12,6 +12,12 @@ This project provides pre-built containers (“toolboxes”) for running LLMs on
* Seamless: shares your home and GPU devices, works like a native shell
* Flexible: easy to switch between Vulkan (open/closed drivers) and ROCm
## 🚨 Updates — 2025-09-28
- Released **ROCm 6.4.4** toolboxes.
- **ROCm-6.4.4+ROCWMMA** is the currently recommended one for most use-cases, but always check the benchmakrs to find the backend that performs better with your model architecture and quantization of choice -> [Performance Benchmarks (Key Results)](#3-performance-benchmarks-key-results)
## Table of Contents
1. [Llama.cpp Compiled for Every Backend](#1-llamacpp-compiled-for-every-backend)
@@ -30,9 +36,6 @@ This project provides pre-built containers (“toolboxes”) for running LLMs on
7. [More Documentation](#7-more-documentation)
8. [References](#8-references)
## 🚨 Updates — 2025-09-28
Released ROCm 6.4.4 toolboxes. ROCm-6.4.4+ROCWMMA is the currently recommenede one for most use-cases, but always check the benchmakrs to find the backend that performs better with your model architecture and quantization of choice -> [Performance Benchmarks (Key Results)](#3-performance-benchmarks-key-results)
## 1. Llama.cpp Compiled for Every Backend