Skip to content

Issues: microsoft/DeepSpeed

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

989 Open 1,739 Closed
989 Open 1,739 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[BUG] ACL stream synchronize failed, error code 507015 bug Something isn't working compression
#6555 opened Sep 19, 2024 by janelu9
CUDA Graphs + ZeRO-3 / TP+PP
#6552 opened Sep 18, 2024 by jeromeku
[BUG] CUDA error: no kernel image is available for execution on the device bug Something isn't working training
#6549 opened Sep 17, 2024 by getao
@loadams
2
[REQUEST] A minimal example to load universal checkpoint enhancement New feature or request
#6548 opened Sep 17, 2024 by hsl89
[BUG] Expert gradient scaling problem with ZeRO optimizer bug Something isn't working training
#6545 opened Sep 17, 2024 by wyooyw
@tohtana
1
RecursionError: maximum recursion depth exceeded while calling a Python object bug Something isn't working training
#6534 opened Sep 13, 2024 by Swordfish1990
@tohtana
2
[REQUEST] dynamic batch size with gradient accumulate enhancement New feature or request
#6533 opened Sep 13, 2024 by Xiang-cd
2
nv-nightly CI test failure ci-failure
#6529 opened Sep 12, 2024 by github-actions bot
[REQUEST] parallelize zero_to_fp32.py to use multiple cpu-cores and threads enhancement New feature or request
#6526 opened Sep 11, 2024 by stas00
[BUG] Distributed Training randomly stuck in trainings loop bug Something isn't working training
#6524 opened Sep 11, 2024 by raeudigerRaeffi
@tohtana
2
[BUG] error ：past_key, past_value = layer_past，how to solve this ? bug Something isn't working deepspeed-chat Related to DeepSpeed-Chat
#6522 opened Sep 11, 2024 by lovychen
1
[BUG] RuntimeError: Error building extension 'inference_core_ops' bug Something isn't working inference
#6519 opened Sep 10, 2024 by Chetan3200
[REQUEST] ZeRO3 doc - support for wrapping model sub-components seperately for training enhancement New feature or request
#6505 opened Sep 8, 2024 by orrzohar
[BUG] Universal checkpointing doesn't work when changing model parallel size (pp and dp change are ok) bug Something isn't working compression
#6503 opened Sep 8, 2024 by exnx
[BUG] Config mesh_device None bug Something isn't working training
#6501 opened Sep 6, 2024 by GuanhuaWang
[BUG] deepspeed tries to call "hostname -I" which is not a valid flag for hostname. it should be "hostname -i" bug Something isn't working training
#6497 opened Sep 5, 2024 by sirus20x6
@loadams
10
[BUG] GatheredParameters resulting in NCCL mismatched collectives error bug Something isn't working training
#6492 opened Sep 5, 2024 by jeromeku
@tohtana
1
Deep Speed Optimizer index out of range during Training bug Something isn't working training
#6482 opened Sep 4, 2024 by manitadayon
@GuanhuaWang
[REQUEST] How to use Flops Profiler to test model.generate(), with customized forward context manager enhancement New feature or request
#6475 opened Sep 2, 2024 by lemyx
Zero3 Partitioning Granularity
#6469 opened Sep 1, 2024 by jeromeku
@tohtana
1
[REQUEST] MiCS vs Zero++ hpZ for Hybrid FSDP enhancement New feature or request
#6467 opened Aug 31, 2024 by jeromeku
5
Flops Profiler
#6466 opened Aug 31, 2024 by BhAem
1
[REQUEST] Replace reduce in ZERO 1/2/3 with reduce_scatter enhancement New feature or request
#6463 opened Aug 30, 2024 by efsotr
8
[BUG] setting PYTORCH_CUDA_ALLOC_CONF in .deepspeed_env raise return code = -11 bug Something isn't working training
#6454 opened Aug 28, 2024 by nomadlx
@tohtana
1
[BUG] Significant difference between using DeepSpeed and not using DeepSpeed bug Something isn't working training
#6451 opened Aug 28, 2024 by hxu105
ProTip! What’s not been updated in a month: updated:<2024-08-19.