Training & fine-tuning
PyTorch 2.5 · CUDA 12.4
Ubuntu 22.04, PyTorch 2.5, CUDA 12.4, cuDNN, JupyterLab, SSH.
NVIDIA GPUsContainer & documentation
spotgpus/pytorch:2.5-cu124Software templates 9 environments
Choose your tools, then find a GPU. From your first notebook to a multi-GPU training job.
9 templates
NVIDIA GPUsDevelopment
CUDA 12.4 base
Ubuntu 22.04, NVIDIA driver, CUDA 12.4 toolkit, Docker, SSH. Bring your own stack.
NVIDIA GPUsContainer & documentation
spotgpus/cuda:12.4
Inference
vLLM inference
vLLM with OpenAI-compatible server on port 8000, Hugging Face cache on the persistent disk.
NVIDIA GPUsContainer & documentation
spotgpus/vllm:latest
Notebooks
JupyterLab
JupyterLab with PyTorch and common data-science packages, token shown in the console.
NVIDIA GPUsContainer & documentation
spotgpus/jupyter:latest
Image generation
ComfyUI
ComfyUI with the manager, models folder on the persistent disk.
NVIDIA GPUsContainer & documentation
spotgpus/comfyui:latest
Inference
Ollama
Ollama server with the API on port 11434, models cached on the persistent disk.
NVIDIA GPUsContainer & documentation
spotgpus/ollama:latest
Training & fine-tuning
TensorFlow 2.17
Ubuntu 22.04, TensorFlow 2.17 GPU, CUDA 12.3, JupyterLab, SSH.
NVIDIA GPUsContainer & documentation
spotgpus/tensorflow:2.17
Development
ROCm 6.2 (AMD)
Ubuntu 22.04, ROCm 6.2, PyTorch for ROCm, SSH. AMD Instinct instances only.
AMD GPUsContainer & documentation
spotgpus/rocm:6.2
Development
Ubuntu 22.04 (driver only)
Plain Ubuntu with the GPU driver and SSH. Nothing else.
General-purpose tools
Container & documentation
spotgpus/ubuntu:22.04No matching templates
Try another search or choose a different category.