Skip to content

Pull requests: microsoft/DeepSpeed

New pull request New
123 Open 2,973 Closed
123 Open 2,973 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix: forbid repeated deepspeed.initialize on training objects
#6874 opened Dec 16, 2024 by traincheck-team Loading…
1
3
Fix --enable_each_rank_log when used with PDSH multi-node runner
#6863 opened Dec 13, 2024 by akeshet Loading…
1
1
Add arctic model support by adding w2 to all_reduce
#6856 opened Dec 11, 2024 by pi314ever Loading…
5
Use ds-specific module id to avoid conflicts
#6847 opened Dec 10, 2024 by tjruwase Loading…
1
[inf] Add config var to enable keeping module on host
#6846 opened Dec 10, 2024 by oelayan7 Loading…
1
Cleanup/fix errors in ops/transformer/inference tests
#6830 opened Dec 6, 2024 by loadams • Draft
Support pure meta model lm_head tp
#6812 opened Dec 2, 2024 by Yejing-Lai Loading…
Zero2: avoid graph breaks in torch.compile by using param_idx
#6803 opened Nov 28, 2024 by nelyahu Loading…
3
Stage3: Use new torch grad accumulation hooks API
#6773 opened Nov 21, 2024 by deepcharm Loading…
5
Check transformers version in BLOOM for inference v1
#6766 opened Nov 19, 2024 by lekurile Loading…
1
BLOOM fixes for DS Legacy Inference
#6765 opened Nov 19, 2024 by lekurile • Draft
Add the missing view operations from sequence parallel(async).
#6750 opened Nov 14, 2024 by inkcherry Loading…
3
Fix building on Windows with presence of Triton
#6749 opened Nov 14, 2024 by woct0rdho Loading…
2
Update flake8 version
#6740 opened Nov 11, 2024 by loadams Loading…
Training ops kernels: Speeding up the Llama-based MoE architectures
#6734 opened Nov 8, 2024 by RezaYazdaniAminabadi • Draft
1
Support latest transformers with DSChat
#6711 opened Nov 4, 2024 by loadams Loading…
Allow launcher to include --include=node3, not just --include=node3:1,2,3,4,5,6,7,8
#6698 opened Nov 1, 2024 by stephen-nju Loading…
1
9
Reduce the device bubble introduced by heavy loop synchronization in coalesced fetch/release(z3_leaf_module)
#6694 opened Oct 31, 2024 by inkcherry Loading…
Update MII tests to support transformers latest
#6686 opened Oct 29, 2024 by loadams Loading…
Allow to compile collective for PT > 2.3
#6674 opened Oct 27, 2024 by nelyahu Loading…
@loadams
3
Support the parallel conversion from ZeRO checkpoints to FP32/FP16/BF16 param weight
#6655 opened Oct 23, 2024 by xylian86 Loading…
5 tasks done
3
[DO NOT MERGE] Log test results to file
#6627 opened Oct 15, 2024 by tohtana • Draft
6
modify_load_save_model
#6626 opened Oct 15, 2024 by ssklzx Loading…
3
Improve consistency of zero_grad
#6554 opened Sep 18, 2024 by tohtana • Draft
Enabled configurable auto Tensor Parallelism (TP) for the inference of diverse models
#6553 opened Sep 18, 2024 by gyou2021 Loading…
4
ProTip! Exclude everything labeled bug with -label:bug.