Skip to content

Pull requests: deepspeedai/DeepSpeed

New pull request New
110 Open 3,106 Closed
110 Open 3,106 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

async tp allreduce
#7115 opened Mar 7, 2025 by inkcherry Loading…
2
[XPU] Support XCCL on deepspeed side
#7113 opened Mar 6, 2025 by ys950902 Loading…
1
fix keep_module_on_host
#7112 opened Mar 6, 2025 by inkcherry Loading…
5
[Draft] Add support for seq split in Domino
#7111 opened Mar 4, 2025 by duanhx1037 • Draft
@GuanhuaWang @duanhx1037
3
Variable batch size and LR scheduler
#7104 opened Mar 3, 2025 by bm-synth Loading…
12
Update gaudi2 nightly,ci to latest 1.20.0 build
#7093 opened Feb 28, 2025 by raza-sikander • Queued
6
Unpin once transformers latest is fixed
#7088 opened Feb 27, 2025 by loadams Loading…
1
Update Domino for Llama3
#7084 opened Feb 26, 2025 by shenzheyu Loading…
1
Conditionally quote env vars
#7071 opened Feb 22, 2025 by saurabhkoshatwar Loading…
1
@loadams
3
Fix, pipeline model with moe cause error when send grad
#7055 opened Feb 19, 2025 by wukong1992 Loading…
3
Enable ZeRO set/get APIs for NVMe offload
#7046 opened Feb 17, 2025 by tjruwase Loading…
9
Add pyproject.toml with legacy build backend to keep most logic in setup.py
#7033 opened Feb 13, 2025 by loadams Loading…
4 of 5 tasks
1
3
Training multiple models
#7018 opened Feb 8, 2025 by tjruwase Loading…
16
Enable python 3.11 and 3.12 tests
#7007 opened Feb 6, 2025 by loadams Loading…
Enable torch.autocast with ZeRO
#6993 opened Feb 3, 2025 by tohtana Loading…
Improve overflow handling in ZeRO
#6976 opened Jan 28, 2025 by tjruwase Loading…
4 of 6 tasks
1
3
Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs
#6964 opened Jan 21, 2025 by gyou2021 Loading…
9
[FPDT] Support FPDT Based on Intel Backend
#6956 opened Jan 16, 2025 by YizhouZ • Draft
Pin numpy version
#6953 opened Jan 15, 2025 by BLOrange-AMD Loading…
3
Set dataloader shuffle=true
#6950 opened Jan 14, 2025 by loadams • Draft
1 task
Update sharded_moe.py to support top2 gate with Tutel
#6948 opened Jan 14, 2025 by xenshinu Loading…
9
[DEBUG] Add diagnostics for cpu-torch-latest intermittent hang
#6942 opened Jan 10, 2025 by loadams • Draft
Fix: forbid repeated deepspeed.initialize on training objects
#6874 opened Dec 16, 2024 by traincheck-team Loading…
1
32
Training ops kernels: Speeding up the Llama-based MoE architectures
#6734 opened Nov 8, 2024 by RezaYazdaniAminabadi • Draft
1
Update MII tests to support transformers latest
#6686 opened Oct 29, 2024 by loadams Loading…
@loadams
ProTip! Updated in the last three days: updated:>2025-03-04.