← All templates

PyTorch (CUDA)

Published, not hardware-testedclassicalpytorch

Approved for the catalog. No GPU run has verified it yet.

Rating
No ratings yet
Runs
0
Min VRAM
8.6 GB
Required backend
CUDA · NVIDIA GPUs
Version
pytorch-cuda-v1.4.0
Image digest
d985a242e7cd…
Readiness
SSH probe on the leased endpoint

SSH-only classical workload; readiness is the SSH endpoint probe (no HTTP readiness). Source: https://github.com/pytorch/pytorch (2.14.0 CUDA 12.6). License: https://github.com/pytorch/pytorch/blob/main/LICENSE including third-party notices; CUDA terms apply.

Compatibility is checked against each GPU's measured runtime and dedicated VRAM. A hardware report or community status does not qualify another vendor backend.

Reported on real GPUs

Users report these themselves. AmpleRun has not verified them.

No hardware reports yet. Be the first to report how it runs on your chip.

Sign in to report how this template runs on your GPU.

Deploy on this template →Shows only machines that fit this template.

Reviews

No reviews yet.

Sign in to leave a review.


Performance data not yet measured: machines are offered by price, not ranked by speed.

Sign in to deploy

Reserving holds funds, so it needs a verified account. The template stays selected through sign-in.