Search
Results
Integrated GPU support · Issue #2637 · ollama/ollama
Latest (0.1.27) docker image with ROCm works for me on Ryzen 5600G with 8GB VRAM allocation. Prompt processing is 2x faster than with CPU. Generation runs at max speed even if CPU is busy running other processes. I am on Fedora 39. Container setup: HSA_OVERRIDE_GFX_VERSION=9.0.0
SDB:AMD GPGPU - openSUSE Wiki
Use this to install ROCm on Tumbleweed. Does NOT talk about Pytorch, the tag is for completeness.
[rocm] No GPU support after rebuild with ROCm 6.0 (#6) · Issues · Arch Linux / Packaging / Packages / python-pytorch · GitLab
This site seems to claim there is a known fix. Looks like someone is fixing it???
Installing PyTorch for ROCm — ROCm installation (Linux)
Installing PyTorch for ROCm - this document claims gfx900 compatibility
Document describing installation. The only supported claim: Ubuntu 22.04.3 Desktop Version with HWE Ubuntu kernel 6.5 Yes
AMD Extends Support for PyTorch Machine Learning Development on Select RDNA™ 3 GPUs with ROCm™ 5.7 | PyTorch
The links in “How to guide“ provide instructions that are hopeful. Maybe start with those instructions!
Doesn't ROCm support AMD's integrated GPU (APU)? · Issue #2216 · ROCm/ROCm
This guy claims successful installation of ROCm on Ubuntu - this seems to be workable for Tumbleweed as well. See the comment “nav9 commented on Jul 16, 2023“
Ditching CUDA for AMD ROCm for more accessible LLM training and inference. | by Rafael Manzano Masapanta | Medium
Train LLM on AMD APU. In this scenario, we’ll use an APU because most laptops with a Ryzen CPU include an iGPU; specifically, this post should work with iGPUs based on the “GCN 5.0” architecture, or “Vega” for friends. We’ll use an AMD Ryzen 2200G in this post, an entry-level processor equipped with 4C/4T and an integrated GPU.