by Hanyuyuan6 · Codex Skill · ★ 52
remote-gpu-trainer An Agent Skill for running long GPU jobs on machines you rent but don't own. Deploy, train, monitor, and tear down safely across AutoDL, RunPod, vast.ai, Lambda, Paperspace, the Chinese platforms (恒源云 / 矩池云 / Featurize / 揽睿星舟), bare SSH boxes, Slurm, and Kubernetes. One instance, or a fan-out of many. Disambiguation: "AutoDL" here is the autodl.com GPU-rental platform, not AutoML or NAS. And this is an Agent Skill — a with reference docs and script templates — not a CLI or an SDK.
| Stars | 52 |
| Forks | 5 |
| Language | Shell |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 66.6514511037527/100 |
| Last Updated | 2026-07-08 |
| Created | 2026-06-18 |
| Platforms | claude-code, cli, codex |
| Est. Tokens | ~14k |
These tools work well together with remote-gpu-trainer for enhanced workflows:
Explore other popular codex skill tools:
remote-gpu-trainer is An Agent Skill for the DL experiment lifecycle: RUN (a GPU you own or rent) → VERIFY the number is real → DELIVER reproducible, single-source figures and tables.. It is categorized as a Codex Skill with 52 GitHub stars.
remote-gpu-trainer is primarily written in Shell. It covers topics such as agent-skills, autodl, claude-code.
You can find installation instructions and usage details in the remote-gpu-trainer GitHub repository at github.com/Hanyuyuan6/remote-gpu-trainer. The project has 52 stars and 5 forks, indicating an active community.
remote-gpu-trainer is released under the MIT license, making it free to use and modify according to the license terms.