# Run ollama on GFX803
* host:ubuntu 22.04
* R5 2600, RX570 4GB(GFX803)

## 測試docker

sudo docker run -it --device=/dev/kfd --device=/dev/dri --group-add=video --ipc=host --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -p 8089:8080 rocm61_pt24:latest /bin/bash

## mozilla llamafile
https://github.com/Mozilla-Ocho/llamafile

wget https://huggingface.co/Mozilla/TinyLlama-1.1B-Chat-v1.0-llamafile/resolve/main/TinyLlama-1.1B-Chat-v1.0.F16.llamafile?download=true

./llava-v1.5-7b-q4.llamafile -ngl 999

解壓缺少的HIP SDK

sudo apt-get update
sudo apt-get install p7zip-full
wget https://github.com/likelovewant/ROCmLibs-for-gfx1103-AMD780M-APU/releases/download/v0.6.1.2/rocm.gfx803.optic.vega10.logic.hip.sdk.6.1.2.7z

7z x rocm.gfx803.optic.vega10.logic.hip.sdk.6.1.2.7z 
cp -r library/* /opt/rocm/lib/rocblas/library/

運行ollama

./llama.llamafile -ngl 999 --host 0.0.0.0

有GPU加速

# 其他補充
Ollama for amd:
https://github.com/likelovewant/ollama-for-amd/wiki