Skip to content
Navigation Menu
Sign in
Products
Platform
GitHub
Repository hosting, code review, and collaboration
Copilot
AI coding agent for tasks, projects, and automated work
Build & Ship
Actions
CI/CD and automation
Code Quality
Quality analysis, fixes, and enforcement
Codespaces
Cloud development environments
Secure
Code Security
Vulnerability detection and remediation
Secret Protection
Secret scanning and leak prevention
All Features
Developers
Learn
Documentation
Training
Certifications
Build
Build with Copilot
Customize Copilot
Integrations
Explore
Trending repos
GitHub Sponsors
Developer articles
Changelog
Changelog
Your source for the latest features, improvements, and fixes across GitHub products, all in one place.
Businesses
By Stage
Enterprises
Team
Startups
Nonprofits
By Industry
Healthcare
Financial services
Manufacturing
Government
See all industries
Evaluate GitHub
Why GitHub
Customer Stories
Business insights
Trust Center
Add-ons
GitHub Advanced Security
Enterprise-grade code security
Copilot for business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Resources
Learn & Connect
Documentation
Resource center
Blog
Events & webinars
Community Forum
Customer Stories
Integrations
Support
Customer support
Training
GitHub Status
Company
About
Partners
Changelog
GitHub Universe
GitHub Universe
The event for developers, by developers — talks, workshops, and what’s next for GitHub.
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
deepspeedai
DeepSpeed
Repository navigation
Code
Issues
1.2k
(1.2k)
Pull requests
374
(374)
Discussions
Actions
Projects
Security and quality
Insights
More
items
All pull requests
Labels
30
(30)
Milestones
0
(0)
New pull request
Search pull requests
is
:
pr
state
:
open
is:pr state:open
Clear filter
Search
Pull requests
Open
374
(374)
Closed
4,241
(4,241)
Author
Label
Projects
Milestones
Reviews
Assignee
Sort by
Newest
descending
Sort…
Sort by Newest descending
More items
Comfortable display density
Compact display density
Fix external loss scaling in TiledFusedLogitsLoss
skip-gpu-test
#9096
·
taking-lying-flat
opened
Nov 16, 2026
Contributor
·
·
1
5
[checkpoint] Prepare an affine transfer once and run it per step
skip-gpu-test
#9094
·
0z5a
opened
Nov 16, 2026
Contributor
·
·
4
[checkpoint] Check aliases on the regions a transfer moves
skip-gpu-test
#9093
·
0z5a
opened
Nov 16, 2026
Contributor
·
·
3
ZeRO-1/2: average gradients in the wider of the gradient and communication dtypes
skip-gpu-test
#9092
·
seanbell
opened
Nov 16, 2026
Contributor
·
·
4
Fix flops profiler reporting 0 MACs for RNN/GRU/LSTM modules and cells
skip-gpu-test
#9090
·
roshankumarg529
opened
Nov 15, 2026
·
·
1
6
Raise when a later engine asks for another tensor parallel layout (#9081 stage 1)
skip-gpu-test
#9088
·
jinyouzhi
opened
Nov 15, 2026
Contributor
·
·
1
[ZeRO-3] Optimize GatheredTensor.contiguous group interval lookup with bisect (#9062)
skip-gpu-test
#9084
·
davrens101
opened
Nov 14, 2026
·
·
3
Gate merge queue GPU CI by PR label
skip-gpu-test
#9083
·
tohtana
opened
Nov 13, 2026
Collaborator
·
·
17
Load universal checkpoints across ZeRO stages
skip-gpu-test
#9080
·
jinyouzhi
opened
Nov 13, 2026
Contributor
·
·
4
Run the whole static rollout decode step inside the CUDA graph
#9079
·
jinyouzhi
opened
Nov 13, 2026
Contributor
·
·
Fix DeepCompile NCCL header discovery for PyTorch wheels
skip-gpu-test
#9078
·
hwchen2017
opened
Nov 13, 2026
Contributor
·
·
1
Reuse the static rollout decode graph across calls
#9077
·
jinyouzhi
opened
Nov 13, 2026
Contributor
·
·
8
[zero] Run ZeRO-3's rank-agreement checks and barriers over the partition group
skip-gpu-test
#9076
·
LiRunGuo
opened
Nov 13, 2026
Contributor
·
·
12
[rollout] In-memory relayout between sharded layouts (two-stage relay, pipelined execution) and ZeRO-3 training-state move
#9074
·
LiRunGuo
opened
Nov 13, 2026
Contributor
·
·
5
write the latest checkpoint tag only after every rank has saved
skip-gpu-test
#9071
·
dundysm
opened
Nov 12, 2026
Contributor
·
·
1
5
[AffineIR] Plan flat shards that keep the parameter's own strides
skip-gpu-test
#9070
·
0z5a
opened
Nov 12, 2026
Contributor
·
·
6
ZeRO-3: take the AutoTP shard from universal checkpoints converted without TP
skip-gpu-test
#9069
·
Achyuthan-S
opened
Nov 12, 2026
Contributor
·
·
3
[Triton] Add Ascend Triton variants and a per-kernel probe policy
#9065
·
VenusTZZ
opened
Nov 12, 2026
Contributor
·
5 tasks done
·
·
Fix replicated K/V all-reduce to run once per accumulation window
skip-gpu-test
#9064
·
VenusTZZ
opened
Nov 12, 2026
Contributor
·
4 tasks done
·
·
Fix flops profiler MAC count for conv with padding='same' and even kernel
skip-gpu-test
#9063
·
davrens101
opened
Nov 12, 2026
·
·
1
Broadcast the mask in Triton inference masked softmax
#9061
·
cervoliu
opened
Nov 11, 2026
·
·
1
8
fix(module-inject): preserve HF precision boundaries in segKI kernels
#9057
·
nathon-lee
opened
Nov 11, 2026
Contributor
·
·
[XPU] Open the Triton gate on Intel XPU
skip-gpu-test
#9055
·
jinyouzhi
opened
Nov 11, 2026
Contributor
·
·
Describe a parameter held in several layouts as one affine map
skip-gpu-test
#9052
·
Achyuthan-S
opened
Nov 11, 2026
Contributor
·
·
7
Fix rank split when a full optimizer state is missing
skip-gpu-test
#9049
·
Shr3yash
opened
Nov 10, 2026
·
·
4
Previous
1
2
3
4
5
6
7
…
15
Next
You can’t perform that action at this time.