consumer
NVIDIA GeForce RTX 4090
Rate $0.44 / hr. Eligible wallets are not charged. Runtime is one hour.
Specs
VRAM
24GB GDDR6X
CUDA cores
16384
Tensor cores
512
Memory bandwidth
1008 GB/s
FP32
82.6 TFLOPS
FP16
165.2 TFLOPS
Used for
Medium training, Image generation, Fine-tuning
Images on this GPU
- PyTorchPyTorch with CUDA, cuDNN, and JupyterLab.pytorch:2.1.0-py3.10-cuda11.8.0-devel-ubuntu22.048888/http 22/tcp
- TensorFlowTensorFlow 2.x with CUDA and JupyterLab.tensorflow/tensorflow:latest-gpu-jupyter8888/http 22/tcp
- Stable DiffusionAutomatic1111 WebUI.stable-diffusion:web-ui-10.2.13000/http 22/tcp
- ComfyUINode-based Stable Diffusion interface.stable-diffusion:comfy-ui-5.1.23000/http 22/tcp
- LLM InferenceText Generation WebUI.pytorch:2.1.0-py3.10-cuda11.8.0-devel-ubuntu22.047860/http 22/tcp
Start
locked
Holders of 10,000 $GPU start against the treasury-funded cluster. Sessions are capped at 1 hour. Everyone else can read specs.
Runtime 1 hour (fixed)
GPU NVIDIA GeForce RTX 4090
Image pytorch:2.1.0-py3.10-cuda11.8.0-devel-ubuntu22.04
Ports 8888/http,22/tcp
Cost ~$0.44 covered if eligible
Connect a wallet on Robinhood Chain.