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

983 Open 1,910 Closed
983 Open 1,910 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

DeepSpeed with ZeRO3 strategy cannot build 'fused_adam' bug Something isn't working training
#6892 opened Dec 18, 2024 by LeonardoZini
How to perform inference MoE model with expert parallel
#6891 opened Dec 18, 2024 by Guodanding
nv-torch-nightly-v100 CI test failure ci-failure
#6888 opened Dec 18, 2024 by github-actions bot
nv-ds-chat CI test failure ci-failure
#6887 opened Dec 18, 2024 by github-actions bot
nv-nightly CI test failure ci-failure
#6883 opened Dec 17, 2024 by github-actions bot
How can DeepSpeed be configured to prevent the merging of parameter groups
#6878 opened Dec 16, 2024 by CLL112
3
How do I know if stage-3 is a success by using deepspeed？ training
#6877 opened Dec 16, 2024 by hwhyyds
1
[BUG] Cannot use --hostfile to start multi-node training in Docker. bug Something isn't working training
#6875 opened Dec 16, 2024 by Ind1x1
@GuanhuaWang
Windows wheel build error - Tried everything with all requirements you have build Improvements to the build and testing systems. windows
#6871 opened Dec 14, 2024 by FurkanGozukara
@loadams
3
[BUG] Invalidate trace cache @ step 10: expected module 11, but got module 19 bug Something isn't working training
#6870 opened Dec 14, 2024 by yafuly
2
[BUG] Mismatch of model parameters when using Sequence Parallel bug Something isn't working training
#6868 opened Dec 13, 2024 by chetwin-character
Unable to Install DeepSpeed on Windows using pip windows
#6865 opened Dec 13, 2024 by H4CK3R-5M4CK3R
@loadams
2
[BUG]When fine-tuning an LLM, the following error occurs after training for some time: self.optimizer.param_groups[param_group_id]['params'] = [] IndexError: list index out of range bug Something isn't working training
#6857 opened Dec 12, 2024 by tdtgi
@jomayeri
2
[BUG] Unable to Use quantization_setting for Customizing MoQ in DeepSpeed Inference bug Something isn't working compression
#6853 opened Dec 11, 2024 by cyx96
@jomayeri
1
DeepSpeed with trl bug Something isn't working training
#6852 opened Dec 11, 2024 by sagie-dekel
@jomayeri
6
[QUESTIONS]：Some questions about running Domino enhancement New feature or request
#6851 opened Dec 11, 2024 by yingtongxiong
1
@GuanhuaWang @hwchen2017
2
Opinion on Refactoring Ulysses enhancement New feature or request
#6843 opened Dec 9, 2024 by Eugene29
Question about Ulysses and loss agregation
#6841 opened Dec 9, 2024 by pavelgein
4
[BUG] inference ops unit tests are failing bug Something isn't working inference
#6839 opened Dec 9, 2024 by oelayan7
@loadams
2
[REQUEST] domino integration to nanotron enhancement New feature or request
#6835 opened Dec 7, 2024 by NouamaneTazi
@GuanhuaWang
zero-3 cpuadam is so slow enhancement New feature or request
#6834 opened Dec 7, 2024 by SeunghyunSEO
3
[BUG] offload optmizer states in zero3 bug Something isn't working training
#6833 opened Dec 7, 2024 by Hanqer
@tohtana
1
[BUG] using deepspeed slower inference time bug Something isn't working inference
#6818 opened Dec 4, 2024 by williamlin0518
[QST] MoE auxiliary loss
#6816 opened Dec 4, 2024 by osayamenja
[BUG] DeepSpeed accuracy issue for torch.compile if activation checkpoint function not compiler disabled bug Something isn't working training
#6811 opened Dec 1, 2024 by NirSonnenschein
ProTip! Updated in the last three days: updated:>2024-12-15.