https://github.com/mixa3607/ml-gfx906
ML software (llama.cpp, ComfyUI, vLLM) builds for AMD gfx906 GPUs, e.g. Radeon VII / MI50 / MI60
https://github.com/mixa3607/ml-gfx906
amd-gpu comfyui llamacpp torch vllm
Last synced: 12 days ago
JSON representation
ML software (llama.cpp, ComfyUI, vLLM) builds for AMD gfx906 GPUs, e.g. Radeon VII / MI50 / MI60
- Host: GitHub
- URL: https://github.com/mixa3607/ml-gfx906
- Owner: mixa3607
- Created: 2025-07-13T20:21:34.000Z (10 months ago)
- Default Branch: master
- Last Pushed: 2026-04-10T23:19:04.000Z (16 days ago)
- Last Synced: 2026-04-11T01:13:58.737Z (16 days ago)
- Topics: amd-gpu, comfyui, llamacpp, torch, vllm
- Language: C#
- Homepage: https://arkprojects.space/wiki/AMD_GFX906
- Size: 13.7 MB
- Stars: 154
- Watchers: 4
- Forks: 21
- Open Issues: 4
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# ML software for deprecated GFX906 arch
## Docs
https://arkprojects.space/wiki/AMD_GFX906
## Prebuild images
### Images
| Name | About | Status | Docs |
| ------------ | -------------------- | ------ | ---------------------------------- |
| ROCm | ROCm patched images | OK | [readme](./rocm/readme.md) |
| ROCm-tensile | gfx906 tensile files | OK | [readme](./rocm-tensile/readme.md) |
| PyTorch | PyTorch images | OK | [readme](./pytorch/readme.md) |
| llama.cpp | llama.cpp images | OK | [readme](./llama.cpp/readme.md) |
| ComfyUI | ComfyUI images | OK | [readme](./comfyui/readme.md) |
| VLLM | vLLM images | OK | [readme](./vllm-v2/readme.md) |
| Project | Image |
| --------- | ----------------------------------------------------------- |
| ROCm | `docker.io/mixa3607/rocm-gfx906:6.3.3-complete` |
| | `docker.io/mixa3607/rocm-gfx906:6.4.4-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.0.0-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.0.2-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.1.0-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.1.1-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.2.0-complete` |
| | `docker.io/mixa3607/rocm-gfx906:7.2.1-complete` |
| --------- | ----------------------------------------------------------- |
| PyTorch | `docker.io/mixa3607/pytorch-gfx906:v2.7.1-rocm-6.4.4` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.7.1-rocm-6.3.3` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.8.0-rocm-6.4.4` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.8.0-rocm-6.3.3` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.9.0-rocm-6.4.4` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.9.0-rocm-6.3.3` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.9.0-rocm-7.0.2` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.10.0-rocm-6.3.3` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.10.0-rocm-7.2.0` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.11.0-rocm-6.3.3` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.11.0-rocm-7.2.0` |
| | `docker.io/mixa3607/pytorch-gfx906:v2.11.0-rocm-7.2.1` |
| --------- | ----------------------------------------------------------- |
| ComfyUI | `docker.io/mixa3607/comfyui-gfx906:v0.19.0-rocm-6.3.3` |
| | `docker.io/mixa3607/comfyui-gfx906:v0.19.0-rocm-7.2.1` |
| --------- | ----------------------------------------------------------- |
| vLLM | `docker.io/mixa3607/vllm-gfx906:0.19.1-rocm-6.3.3-aiinfos` |
| | `docker.io/mixa3607/vllm-gfx906:0.19.1-rocm-7.2.1-aiinfos` |
| --------- | ----------------------------------------------------------- |
| llama.cpp | `docker.io/mixa3607/llama.cpp-gfx906:b8799-rocm-6.3.3` |
| | `docker.io/mixa3607/llama.cpp-gfx906:b8799-rocm-7.2.1` |
### Deps graph
```mermaid
flowchart TD
rocm-src[docker.io/rocm/dev-ubuntu-24.04] --> rocm[docker.io/mixa3607/rocm-gfx906]
rocm --> llama[docker.io/mixa3607/llama.cpp-gfx906]
rocm --> torch[docker.io/mixa3607/pytorch-gfx906]
torch --> comfyui[docker.io/mixa3607/comfyui-gfx906]
torch --> vllm[docker.io/mixa3607/vllm-gfx906]
```
## RVS
```shell
apt update && apt install -y rocm-validation-suite
echo 'actions:
- name: gst-581Tflops-4K4K8K-rand-bf16
device: all
module: gst
log_interval: 10000
ramp_interval: 5000
duration: 120000
hot_calls: 1000
copy_matrix: false
target_stress: 581000
matrix_size_a: 4864
matrix_size_b: 4096
matrix_size_c: 8192
matrix_init: rand
data_type: bf16_r
lda: 8320
ldb: 8320
ldc: 4992
ldd: 4992
transa: 1
transb: 0
alpha: 1
beta: 0' > ~/gst-581Tflops-4K4K8K-rand-bf16.conf
/opt/rocm/bin/rvs -c ~/gst-581Tflops-4K4K8K-rand-bf16.conf
```