Back to skills
SkillHub ClubAnalyze Data & AIFull StackData / AI

pytorch-fsdp2

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

Packaged view

This page reorganizes the original catalog entry around fit, installability, and workflow context first. The original raw source lives below.

Stars
5,243
Hot score
99
Updated
March 20, 2026
Overall rating
C4.5
Composite score
4.5
Best-practice grade
B75.6

Install command

npx @skill-hub/cli install zechenzhangagi-ai-research-skills-pytorch-fsdp2
PyTorchFSDP2Fully Sharded Data ParallelDistributed TrainingDTensorDevice MeshSharded CheckpointingMixed PrecisionOffloadTorch Distributed

Repository

zechenzhangAGI/AI-research-SKILLs

Skill path: 08-distributed-training/pytorch-fsdp2

Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.

Open repository

Best for

Primary workflow: Analyze Data & AI.

Technical facets: Full Stack, Data / AI.

Target audience: Development teams looking for install-ready agent workflows..

License: Unknown.

Original source

Catalog source: SkillHub Club.

Repository owner: zechenzhangAGI.

This is still a mirrored public skill entry. Review the repository before installing into production workflows.

What it helps with

  • Install pytorch-fsdp2 into Claude Code, Codex CLI, Gemini CLI, or OpenCode workflows
  • Review https://github.com/zechenzhangAGI/AI-research-SKILLs before adding pytorch-fsdp2 to shared team environments
  • Use pytorch-fsdp2 for development workflows

Works across

Claude CodeCodex CLIGemini CLIOpenCode

Favorites: 0.

Sub-skills: 0.

Aggregator: No.

pytorch-fsdp2 | SkillHub