deepspeedai/DeepSpeedExamples

View on GitHub

Example models using DeepSpeed

Stars: 6.8k
Forks: 1.1k
Open beginner issues: 0
Indexed issues: 300
Avg merge: 2d 16h
Merged PRs (30d): 1

Dominant language: Python
License: Apache-2.0
Last GitHub push: Aug 26, 2026
Latest indexed: Sep 20, 2026
Contributing guide: No contributing guide
Code of conduct: Code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

300 open issues indexed

Question: Why did you implemented LoRA on your hand instead of using peft? Open

Difficulty 3/5 1-2 days Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#830 · 1 comment · Opened Dec 13, 2023
deeepspeed chat 支持pipline 并行吗？ Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#825 · 1 reaction · Opened Dec 5, 2023
Something wrong at step1_supervised_finetuning/main.py Open

Difficulty 2/5 1-3 hours Newbie friendliness 55/100

deepspeedai/DeepSpeedExamples#824 · Opened Dec 5, 2023
Should it use global_rank as the condition for shared-disk? Open

Difficulty 2/5 1-3 hours Newbie friendliness 38/100

deepspeedai/DeepSpeedExamples#822 · Opened Dec 2, 2023
[BUG] DeepSpeed-Chat Step3 - actor model repeats generating the same token when hybrid engine enabled Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#821 · 9 comments · Opened Nov 30, 2023
Why is the shape of rm model all 0 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#820 · 2 comments · Opened Nov 29, 2023
DeepSpeed-VisualChat Tensor shape mismatch Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#818 · Opened Nov 27, 2023
step3 use same memory when I increase GPUs Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#817 · 1 comment · Opened Nov 24, 2023
Does the DeepSpeedVisualChat model have the capability to locate targets, such as generating coordinates for bounding box positions? Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#816 · Opened Nov 23, 2023
Llama2 as actor using zero_stage3 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#814 · 1 comment · Opened Nov 18, 2023
DeepSpeed-Chat Step-1 training error Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#813 · 1 comment · Opened Nov 17, 2023
DeepSpeed-Chat Step-3 tensorboard loss figures with multiple training epochs Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#811 · Opened Nov 16, 2023
No ignore index Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#809 · Opened Nov 12, 2023
Why does the rw_eval.py script return a different score for the same sample each time? Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#808 · 13 comments · Opened Nov 11, 2023
Why does the chat.py script model answer normally, but answer repeatedly when step3 Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#807 · Opened Nov 10, 2023
NCCL watchdog thread terminated with exception Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#803 · Opened Nov 7, 2023
Inquiry about Embedding Concatenation in DeepSpeed-VisualChat Open

deepspeedai/DeepSpeedExamples#800 · 4 comments · 1 reaction · 1 assignee · Opened Nov 6, 2023
Question about loading Dahous dataset from local path. Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#796 · 9 comments · Opened Nov 1, 2023
Why is the PPL so high in the beginning of Step-1 (SFT)? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#793 · Opened Oct 28, 2023
No response when running deepspeed-chat Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#792 · 1 comment · Opened Oct 27, 2023
How to save memory during inference Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#791 · 1 comment · Opened Oct 27, 2023
Error Encountered When Running pipeline_parallelism in deepspeedexample Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#790 · 4 comments · Opened Oct 27, 2023
No such file or directory: '/tmp/data_files/traindata_04ec1ee09de8642fa1c4b659535fca5dca42643897e6a76769bb674209ad4704.pt' Open

Difficulty 4/5 3-5 days Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#787 · 2 comments · Opened Oct 23, 2023
step 3 "run_6.7b_lora.sh" doesn't work with a100 80GB single gpu. Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#786 · 1 comment · Opened Oct 23, 2023
Overflow in deepspeed-chat LoRA and BF16 mode Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#783 · 1 comment · Opened Oct 20, 2023
Wrong import in inference quantization example Open

Difficulty 1/5 Under an hour Newbie friendliness 45/100

deepspeedai/DeepSpeedExamples#774 · 1 comment · Opened Oct 17, 2023
in place Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#771 · Opened Oct 16, 2023
Missing dependence in requirement.txt of applications/DeepSpeed-VisualChat Open

Difficulty 1/5 Under an hour Newbie friendliness 55/100

deepspeedai/DeepSpeedExamples#770 · Opened Oct 16, 2023
"enable_mixed_precision_lora" just available for fp16 tensor? Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#768 · Opened Oct 13, 2023
generation too slow of actor model during generate_experience in step3 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#764 · 3 comments · Opened Oct 11, 2023
run step3 66b opt report an error out of memory Open

deepspeedai/DeepSpeedExamples#763 · 3 comments · 1 assignee · Opened Oct 10, 2023
Why not just use zero3 inference to generate sequence in DeepSpeed Chat stage3 training? Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

deepspeedai/DeepSpeedExamples#760 · 3 comments · Opened Oct 8, 2023
Question about `compute_rewards` in DeepSpeedChat Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#751 · 3 reactions · Opened Oct 2, 2023
indices should be either on cpu or on the same device as the indexed tensor (cpu) Open

Difficulty 4/5 3-5 days Newbie friendliness 18/100

deepspeedai/DeepSpeedExamples#743 · Opened Sep 27, 2023
Same model Llama 7B, why does zero3 initialize different parameter sizes？ Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#742 · 1 comment · Opened Sep 27, 2023
DeepSpeedChat step 1 can not train on big datasets that need to be stored on disk and loaded by the dataloader Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#740 · 1 reaction · Opened Sep 23, 2023
in step 2, I download AdamG012 / chat-opt-350m-reward-deepspeed and run eval_rw.py using it, but got strange reward Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deepspeedai/DeepSpeedExamples#737 · 2 comments · Opened Sep 20, 2023
About available pinned memory limit Open

Difficulty 3/5 1-2 days Newbie friendliness 32/100

deepspeedai/DeepSpeedExamples#732 · Opened Sep 14, 2023
Resizing model embedding when loading the model Open

deepspeedai/DeepSpeedExamples#728 · 3 comments · 1 reaction · 1 assignee · Opened Sep 11, 2023
Inconsistencies in DeepSpeed config: Impact on training and potential issues Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#725 · Opened Sep 9, 2023
[QUESTION] Why Change Special Tokens ID After Training DeepSpeedChat Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#724 · Opened Sep 9, 2023
Support for Q-LoRA Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#723 · Opened Sep 8, 2023
Unable to inference Falcon-40b with deepspeed. Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#720 · 1 reaction · Opened Sep 7, 2023
[PROBLEM] DeepSpeedChat Create HF Model FOR LLAMA Token ID Question Open

Difficulty 5/5 Over a week Newbie friendliness 18/100

deepspeedai/DeepSpeedExamples#719 · 1 comment · Opened Sep 5, 2023
ppo zero stage 3 generate too slow Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#718 · 1 comment · Opened Sep 3, 2023
[QUESTION]I have a question about deepspeed compression on weight quantization and activation quantization Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#716 · Opened Sep 1, 2023
I have a question about zeroquant Example Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#715 · Opened Sep 1, 2023
The problem of model parallelism in training Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

deepspeedai/DeepSpeedExamples#708 · 1 comment · Opened Aug 29, 2023
LLaMA2-7b error:AssertionError: {'id': 292, 'status': 'NOT_AVAILABLE', 'numel': 0, 'ds_numel': 0, 'shape': (0,), 'ds_shape': (0, 4096), 'requires_grad': True, 'grad_shape': None, 'persist': True, 'active_sub_modules': {453}, 'ds_tensor.shape': torch.Size([0])} Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#707 · 7 comments · Opened Aug 29, 2023
Step3: 8 * A100-40G training LLAMA2-7B and OPT-350M out of memory. Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

deepspeedai/DeepSpeedExamples#705 · 2 comments · Opened Aug 25, 2023

deepspeedai/DeepSpeedExamples

Get new issues in your inbox