home/categories/framework-internals/orchestra-research-ai-research-skills-08-distributed-training-pytorch-fsdp2-skill-md
framework-internalsdevelopment

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.

Orchestra-Research
maintainer
Orchestra-Research
اپ ڈیٹ ہوا 1/30/2026
اسٹارز
6563
فورکس
515
quick start

Installation and usage

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.

انسٹالیشن
$ install --globalskills.sh
استعمال

انسٹال کرنے کے بعد، آپ یہ اسکل ٹرمینل میں درج ذیل کمانڈ چلا کر استعمال کر سکتے ہیں:

skills use pytorch-fsdp2