-
Notifications
You must be signed in to change notification settings - Fork 1.3k
Pull requests: huggingface/accelerate
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix KMP_AFFINITY incorrectly set for non-CPU training
#3912
opened Jan 23, 2026 by
hexfaker
Loading…
fix bug when both cpu_ram_efficient_loading and cpu_offload are enabled
#3910
opened Jan 20, 2026 by
kaixuanliu
Loading…
FSDP2: Activation-checkpoint modules directly matching auto-wrap policy, not those whose parent module matches policy
#3909
opened Jan 17, 2026 by
excepshenal
Loading…
3 of 5 tasks
Fix typo in mult-node gpu example script
#3906
opened Jan 13, 2026 by
timonmerk
Loading…
1 of 5 tasks
Support stateful_dataloader for IterableDatasetShard
#3891
opened Dec 19, 2025 by
jcppython
Loading…
2 of 5 tasks
enhance new codes to XPU, and make them be device agnostic
#3890
opened Dec 18, 2025 by
yao-matrix
Loading…
(fix) remove sampler_is_batch_sampler code in prepare_data_loader(..)
contributions-welcome
wip
Work in progress
#3469
opened Mar 31, 2025 by
suzyahyah
Loading…
optimize infer_auto_device_map for multi-GPU allocation
wip
Work in progress
#3321
opened Jan 2, 2025 by
Nech-C
Loading…
3 of 5 tasks
handle weight sharing with init_on_device
wip
Work in progress
#2737
opened May 3, 2024 by
aws-rhsoln
Loading…
5 tasks
[DO NOT MERGE] add all level buffer support when computing infer_auto_device_map
enhancement
New feature or request
feature request
Request for a new feature to be added to Accelerate
wip
Work in progress
#2663
opened Apr 12, 2024 by
SunMarc
Loading…
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.