https://github.com/jalberty2018/run-pytorch-cuda-develop
Compile environment for Pytorch with CUDA
https://github.com/jalberty2018/run-pytorch-cuda-develop
cloud code-server compiler cuda cuda-toolkit docker-image flash-attn jupyterlab python python3 pytorch sage-attention
Last synced: 23 days ago
JSON representation
Compile environment for Pytorch with CUDA
- Host: GitHub
- URL: https://github.com/jalberty2018/run-pytorch-cuda-develop
- Owner: jalberty2018
- Created: 2025-07-05T14:54:04.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2026-03-10T17:07:12.000Z (2 months ago)
- Last Synced: 2026-03-10T19:34:53.374Z (2 months ago)
- Topics: cloud, code-server, compiler, cuda, cuda-toolkit, docker-image, flash-attn, jupyterlab, python, python3, pytorch, sage-attention
- Language: Python
- Homepage:
- Size: 69.3 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
[](https://hub.docker.com/r/ls250824/run-pytorch-cuda-develop)
# run-pytorch-cuda-develop
## Synopsis
A streamlined setup for developing on high-performance hardware.
- Authentication credentials set via secrets for:
- **Code server** authentication
- **Jupyterlab** authentication
- **Hugging Face** and **CivitAI** tokens for model access (mandatory).
Ensure that the required environment variables and secrets are correctly set before running the pod.
See below for options.
## Hardware provisioning
- [Runpod.io](https://runpod.io/)
- GPU
## Software Repositories
### Core
- [Code Server](https://github.com/coder/code-server)
- [Jupyterlab](https://jupyter.org)
- [Jupyter Server](https://jupyter-server.readthedocs.io/en/latest/index.html)
- [HuggingFace cli](https://huggingface.co/docs/huggingface_hub/guides/cli)
- [Nvidia CUDA](https://hub.docker.com/r/nvidia/cuda/tags?name=12)
- [Pytorch.org](https://pytorch.org)
- [Triton](https://triton-lang.org/main/index.html)
## Documentation
- [SM architecture](https://developer.nvidia.com/cuda-gpus)
- [Wikipedia](https://en.wikipedia.org/wiki/CUDA)
## Available Images
### Pytorch 2.8 CUDA 12.9
```bash
docker pull ls250824/run-pytorch-cuda-develop:09102025
```
### Pytorch 2.9 CUDA 13.0
```bash
docker pull ls250824/run-pytorch-cuda-develop:05112025
```
### Pytorch 2.9 CUDA 12.8
```bash
docker pull ls250824/run-pytorch-cuda-develop:08112025
```
### Pytorch 2.10 CUDA 12.8.1
```bash
docker pull ls250824/run-pytorch-cuda-develop:10032026
```
## Settings
### Services
| Service | Port |
|-----------------|---------------|
| **Code Server** | `9000` (HTTP) |
| **Jupyterlab** | `8888` (HTTP) |
| **SSH/SCP** | `22` (TCP) |
| **Gradio** | `7860` (HTTP) |
### Authentication Tokens
| Token | Environment Variable |
|--------------|----------------------|
| Civitai | `CIVITAI_TOKEN` |
| Huggingface | `HF_TOKEN` |
| Code Server | `PASSWORD` |
| Jupyterlab | `JUPYTERLAB_PASS` |
### Sources
| Variable | Description |
|------------------|----------------------------------|
| `SOURCE[1-50]` | source download links (compressed or plain) |
## Building the Docker Image
### `build-docker.py` script options
| Option | Description | Default |
|----------------|-----------------------------------------------------|------------------------|
| `--username` | Docker Hub username | Current user |
| `--tag` | Tag to use for the image | Today's date |
| `--latest` | If specified, also tags and pushes as `latest` | Not enabled by default |
### Build & push Command
Run the following command to clone the repository and build the image:
```bash
git clone https://github.com/jalberty2018/run-pytorch-cuda-develop.git
cp run-pytorch-cuda-develop/build-docker.py ..
export DOCKER_BUILDKIT=1
export COMPOSE_DOCKER_CLI_BUILD=1
python build-docker.py \
--username= \
--tag= \
run-pytorch-cuda-develop
```
Note: If you want to push the image with the latest tag, add the --latest flag at the end.