transformers
Refactor core_model_loading to support FSDP shard-on-read loading
#44974
Open

Refactor core_model_loading to support FSDP shard-on-read loading #44974

3outeille wants to merge 3 commits into fsdp-vs-ddp from fsdp-core-model-loading
3outeille
3outeille 3outeille force pushed from e5fc7eb3 to 4b2a9216 56 days ago
HuggingFaceDocBuilderDev
3outeille 3outeille force pushed from d48fcc7e to 607cc114 54 days ago
3outeille
3outeille commented on 2026-04-07
3outeille 3outeille force-pushed the fsdp-vs-ddp branch from 978ac872 to a5c25548 36 days ago
3outeille DistributedConfig + shard-on-read loading
739332cd
3outeille 3outeille force pushed from 607cc114 to 739332cd 36 days ago
3outeille 3outeille force-pushed the fsdp-vs-ddp branch from 864e9fa4 to 7f6cd3d8 35 days ago
3outeille 3outeille force pushed from dbc96197 to c5672400 35 days ago
3outeille 3outeille force-pushed the fsdp-vs-ddp branch from 7f6cd3d8 to 37dcc14d 35 days ago
3outeille Merge branch 'fsdp-vs-ddp' into fsdp-core-model-loading
c1dab9eb
3outeille 3outeille force pushed from c5672400 to c1dab9eb 35 days ago
3outeille Fix ruff formatting in core_model_loading.py
21f05610
github-actions

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone