by saforem2 · Agent Tool · ★ 34
🍋 ezpz Write once, run anywhere. makes distributed PyTorch launches portable across any supported hardware {NVIDIA, AMD, Intel, MPS, CPU} with zero code changes. Features Multi-hardware — automatic device detection and backend selection (CUDA/NCCL, XPU/CCL, MPS, CPU/Gloo) Zero code changes — same script runs on a laptop, a single GPU, or a thousand-node supercomputer HPC integration — native PBS and SLURM support with automatic hostfile discovery and rank assignment Metric tracking — built-in class for recording, plotting, and saving training metrics CLI tools — , , for launching jobs,...
| Stars | 34 |
| Forks | 8 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 61.8894069602466/100 |
| Open Issues | 2 |
| Last Updated | 2026-07-06 |
| Created | 2023-09-12 |
| Platforms | python |
| Est. Tokens | ~14k |
Explore other popular agent tool tools:
ezpz is Write once, run anywhere; ezpz 🍋. It is categorized as a Agent Tool with 34 GitHub stars.
ezpz is primarily written in Python. It covers topics such as ai-tools, deepspeed, distributed-training.
You can find installation instructions and usage details in the ezpz GitHub repository at github.com/saforem2/ezpz. The project has 34 stars and 8 forks, indicating an active community.
ezpz is released under the MIT license, making it free to use and modify according to the license terms.