Skip to content

Pull requests: microsoft/DeepSpeed

New pull request New
143 Open 2,653 Closed
143 Open 2,653 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

Disable nvtx decorator to avoid graph break
#5697 opened Jun 25, 2024 by tohtana Loading…
sequence parallel with communication overlap
#5691 opened Jun 21, 2024 by inkcherry Loading…
ENV var added for recaching in INF Unit tests
#5688 opened Jun 20, 2024 by raza-sikander Loading…
2
inference unit test injectionPolicy split world_size to multiple tests
#5687 opened Jun 20, 2024 by oelayan7 Loading…
2
Switch from torch.cuda.amp.custom_fwd to torch.amp.custom_fwd(device=...)
#5684 opened Jun 18, 2024 by loadams • Draft
Switch what versions of python are supported
#5676 opened Jun 17, 2024 by loadams • Draft
[CPU] add fp16 support to shm inference_all_reduce
#5669 opened Jun 17, 2024 by delock • Queued
4
Update xpu-max1100.yml with new config and add some tests
#5668 opened Jun 17, 2024 by Liangliang-Ma Loading…
6
Add and Remove ZeRO 3 Hooks
#5658 opened Jun 13, 2024 by jomayeri Loading…
4
Unpin transformers version
#5650 opened Jun 12, 2024 by loadams Loading…
reduce all-to-all communication volume when both expert and non-expert are tensor-parallel
#5626 opened Jun 7, 2024 by taozhiwei Loading…
11
Hybrid Offloading for ZeRO3
#5625 opened Jun 7, 2024 by tohtana • Draft
fix: quantization with DeepSpeed HE
#5624 opened Jun 6, 2024 by Atry Loading…
1
2
Add support for Phi-3 small to FastGen
#5614 opened Jun 4, 2024 by adk9 • Draft
[INF] Enable torch compile for inference
#5612 opened Jun 4, 2024 by oelayan7 Loading…
6
Upgrade HPU image to v1.16.2.
#5610 opened Jun 4, 2024 by vshekhawat-hlab Loading…
5
Add an argument to enable the injection of missing state during the conversion of universal checkpoints
#5608 opened Jun 3, 2024 by xylian86 Loading…
[CPU] Allow deepspeed.comm.inference_all_reduce in torch.compile graph
#5604 opened Jun 3, 2024 by delock Loading…
2
state_dict_factory: llama checkpoint - support SWIGLU
#5601 opened Jun 2, 2024 by nelyahu Loading…
2
FastGen H100 MoE support: Add PyTorch multi-gemm MOE implementation
#5586 opened May 29, 2024 by HeyangQin Loading…
7
Update profiler.py
#5584 opened May 29, 2024 by gameofdimension Loading…
reduce cpu host overhead when using moe
#5578 opened May 29, 2024 by ranzhejiang Loading…
7
Reuse KV cache of prefixes
#5572 opened May 27, 2024 by tohtana • Draft
3
Add support for Microsoft Phi-3 model to DeepSpeed-FastGen
#5559 opened May 21, 2024 by adk9 Loading…
Add chatglm2 & chatglm3 autotp
#5540 opened May 16, 2024 by Yejing-Lai Loading…
4
ProTip! Mix and match filters to narrow down what you’re looking for.