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 }}
NVIDIA
Megatron-LM
Repository navigation
Code
Issues
518
(518)
Pull requests
1.6k
(1.6k)
Discussions
Actions
Projects
Security and quality
Insights
More
items
All pull requests
Labels
105
(105)
Milestones
2
(2)
New pull request
Search pull requests
is
:
pr
state
:
open
is:pr state:open
Clear filter
Search
Pull requests
Open
1,649
(1,649)
Closed
5,785
(5,785)
Author
Label
Projects
Milestones
Reviews
Assignee
Sort by
Newest
descending
Sort…
Sort by Newest descending
More items
Comfortable display density
Compact display density
Only access torch.jit.script on PyTorch < 2.2
complexity: low
determinism-exempt
Final Review
#9631
·
janbernloehr
opened
Nov 20, 2026
Contributor
·
3 of 7 tasks
·
·
5
[Dev] fix(offload): keep offload sources that record_stream cannot protect until copied
complexity: low
#9630
·
buptzyb
opened
Nov 20, 2026
Contributor
·
7 tasks done
·
·
8
[MFSDP] Centralize gradient accumulation and view installation
complexity: low
#9629
·
wujingyue
opened
Nov 20, 2026
Contributor
·
·
1
Keep shortcut MoE overlap collectives on the main stream
#9627
·
santhnm2
opened
Nov 20, 2026
Contributor
·
·
1
fix(ci): accept completed Testmon baseline producers on dev
#9626
·
balasaajay
opened
Nov 20, 2026
Contributor
·
·
1
[training] Add native config-only pretraining entrypoint
#9625
·
cuichenx
opened
Nov 20, 2026
Contributor
·
3 of 7 tasks
·
·
1
ci: isolate MFSDP divergence in cuBLASLt across GB200 and GB300
#9624
·
wujingyue
opened
Nov 20, 2026
Contributor
·
·
6
ci: pack small-world unit tests onto the runner and trim #8426's CI time
#9623
·
lmcafee-nvidia
opened
Nov 20, 2026
Contributor
·
1 of 6 tasks
·
·
1
[GTP] Use native TE subclass dispatch for GTP checkpoints
complexity: low
Final Review
#9622
·
fanshiqing
opened
Nov 20, 2026
Member
·
5 of 7 tasks
·
·
1
6
Co-activation expert placement for the all-to-all dispatcher, with route capture
community-request
#9621
·
RadhaGulhane13
opened
Nov 20, 2026
·
7 tasks
·
·
1
MoE token dedup for the all-to-all dispatcher
community-request
#9620
·
RadhaGulhane13
opened
Nov 20, 2026
·
7 tasks
·
·
1
Add Nemotron Labs 3 Puzzle HybridModel training example
#9619
·
Phlip79
opened
Nov 20, 2026
Member
·
1 task
·
·
1
Test gradients and peak CUDA memory for wide residual and mHC replay
complexity: low
#9618
·
kwangjunahn
opened
Nov 19, 2026
Contributor
·
·
8
fix(ci): select Testmon baselines with published caches
Approved
complexity: low
#9616
·
balasaajay
opened
Nov 19, 2026
Contributor
·
1 task
·
·
1
Fix Gigatoken chat encoding with Transformers 5.23
Approved
complexity: low
#9613
·
cuichenx
opened
Nov 19, 2026
Contributor
·
3 of 7 tasks
·
·
3
build: prepare bridge extra and package discovery
#9612
·
balasaajay
opened
Nov 19, 2026
Contributor
·
4 tasks
·
·
2
test(fsdp): add DSv3 proxy long convergence coverage
#9611
·
wujingyue
opened
Nov 19, 2026
Contributor
·
·
1
MRL: Allow multiple groups to assemble in parallel
Approved
complexity: low
#9609
·
tdene
opened
Nov 19, 2026
Contributor
·
1 of 7 tasks
·
·
4
Port Bridge Llama, GPT-OSS and DeepSeek-V3 conversion into MLM
#9608
·
kamran-nvidia
opened
Nov 19, 2026
Contributor
·
4 of 7 tasks
·
·
1
Fix double-counted packed-sequence FLOP stats with the sequence-packing scheduler
#9604
·
dingqingy-nv
opened
Nov 19, 2026
Contributor
·
1 task
·
·
1
Muon: resharded distributed Newton-Schulz for wide GTP-only weights (…
#9603
·
wanyingw
opened
Nov 19, 2026
Contributor
·
7 tasks
·
·
1
Add DSv4.1-Flash optimizer setup
26.12
#9601
·
janEbert
opened
Nov 19, 2026
Contributor
·
1 task
·
·
1
2
Unify CP layout APIs and module boundaries on the canonical backend
#9599
·
FDecaYed
opened
Nov 19, 2026
Contributor
·
2 of 7 tasks
·
·
1
Inference: Enable MXFP8 Shortcut Overlap
complexity: low
Final Review
#9598
·
mathemakitten
opened
Nov 19, 2026
Contributor
·
1 of 7 tasks
·
·
4
[otel 12/12] Report structured GPU-sniff measurements through the telemetry shim
complexity: medium
#9597
·
ahmadki
opened
Nov 19, 2026
Member
·
6 of 7 tasks
·
·
1
Previous
1
2
3
4
5
6
7
…
40
Next
You can’t perform that action at this time.