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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[BUG] 'DeepSpeedGPTInference' object has no attribute 'dtype' for int8 inference
bug
Something isn't working
inference
#4813
opened Dec 14, 2023 by
jxysoft
[BUG]RuntimeError: Called with more columns than supported, please report this bug and this limit will be increased."
bug
Something isn't working
inference
#4810
opened Dec 13, 2023 by
zhenlu0320
[BUG] Deepspeed Zero 3 Inference InFlight Params with new HuggingFace Mixtral Model
bug
Something isn't working
inference
#4808
opened Dec 13, 2023 by
ryandeng1
[BUG]deepspeed always load the whole model to each gpu, then OOM
bug
Something isn't working
training
#4807
opened Dec 13, 2023 by
aohan237
[BUG] Zero2/3 segmentation fault with CPU optimizer off-loading
bug
Something isn't working
training
#4802
opened Dec 12, 2023 by
haixpham
[REQUEST] what‘s the difference of pipeline Parallelism between deepspeed and megatron?
enhancement
New feature or request
#4801
opened Dec 12, 2023 by
mollon650
DeepSpeed CI failures in Transformers with the latest version 0.12.4 but works with 0.12.3
ci-failure
#4795
opened Dec 11, 2023 by
pacman100
How to inference with data parallelism and model parallelism[BUG]
bug
Something isn't working
inference
#4794
opened Dec 11, 2023 by
HackGiter
Importing DeepSpeed causes a ResourceWarning on closing Python [BUG]
bug
Something isn't working
training
#4793
opened Dec 9, 2023 by
rosario-purple
[BUG] Hybrid Engine with DeepSpeed Stage 3 results and Llama V2 results in gibberish outputs
bug
Something isn't working
deepspeed-chat
Related to DeepSpeed-Chat
#4788
opened Dec 8, 2023 by
pacman100
can deepspeed fast-gen support int8 weightonly inference
enhancement
New feature or request
#4786
opened Dec 8, 2023 by
liting6259
[BUG] The forward hook function of Zero stege3 might misjudge the current step of the training process whether it is in the forward computation or in the backward recomputation
bug
Something isn't working
training
#4784
opened Dec 7, 2023 by
henryhe4004
[BUG] Failed to checkpoint with deepspeed 0.12.4
bug
Something isn't working
training
#4781
opened Dec 7, 2023 by
imoneoi
[BUG] Can't perform gradient accumulation on pipeline parallelism with inputs of different lengths
bug
Something isn't working
training
#4777
opened Dec 6, 2023 by
bm-synth
Deepspeed fails with frozen weights (e.g. only train llama2 embedding layer)
bug
Something isn't working
training
#4776
opened Dec 5, 2023 by
rucnyz
[Question]Significant differences between deepspeed and torchrun training results
bug
Something isn't working
training
#4762
opened Dec 3, 2023 by
getao
[BUG] [ERROR] [autotuner.py:699:model_info_profile_run] The model is not runnable with DeepSpeed with error = (
bug
Something isn't working
training
#4759
opened Dec 2, 2023 by
yongjer
The training parameters have not changed
bug
Something isn't working
training
#4758
opened Dec 2, 2023 by
191220042
Can DeepSpeed support Keras models based on the Torch backend?[REQUEST]
enhancement
New feature or request
#4754
opened Nov 30, 2023 by
pass-lin
Build issues on ROCm with random_ltd extension
rocm
AMD/ROCm/HIP issues
#4753
opened Nov 29, 2023 by
Hobbes-Le-Chat
Where can I find MCR-DL?
enhancement
New feature or request
#4751
opened Nov 29, 2023 by
mayank31398
Previous Next
ProTip!
no:milestone will show everything without a milestone.

