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
TransformerEngine
Repository navigation
Code
Issues
162
(162)
Pull requests
245
(245)
Discussions
Actions
Security and quality
Insights
More
items
All pull requests
Labels
81
(81)
Milestones
2
(2)
New pull request
Search pull requests
is
:
pr
state
:
open
is:pr state:open
Clear filter
Search
Pull requests
Open
245
(245)
Closed
2,806
(2,806)
Author
Label
Projects
Milestones
Reviews
Assignee
Sort by
Newest
descending
Sort…
Sort by Newest descending
More items
Comfortable display density
Compact display density
[JAX] Support Python 3.14 in the MNIST example
org-contribution
#3896
·
aybchan
opened
Nov 18, 2026
Member
·
8 of 13 tasks
·
·
1
[Common] Add CuTeDSL MXFP8 grouped scaled swiglu kernel
#3895
·
kainzhong
opened
Nov 18, 2026
Collaborator
·
13 tasks
·
·
[JAX] MoEBlock aware of Dense TP axes (compound EP)
#3894
·
tdophung
opened
Nov 18, 2026
Collaborator
·
8 of 13 tasks
·
·
fix(permute): do not backpropagate through a zero-probability routed row
community-contribution
#3893
·
lxcxjxhx
opened
Nov 18, 2026
Contributor
·
·
1
2
fix(unpermute): infer the token count from the row map when
restore_shape
is omitted
community-contribution
#3892
·
lxcxjxhx
opened
Nov 18, 2026
Contributor
·
·
1
1
fix(cross_entropy): keep the denominator finite when a block is entirely
-inf
community-contribution
#3891
·
lxcxjxhx
opened
Nov 18, 2026
Contributor
·
·
1
7
[Common] Add CuTeDSL MXFP8 group dequantize kernel
#3890
·
kainzhong
opened
Nov 18, 2026
Collaborator
·
13 tasks
·
·
docs: how to enable comm+GEMM overlap in your own model
community-contribution
#3889
·
lxcxjxhx
opened
Nov 18, 2026
Contributor
·
·
13
Bump nccl ep submodule
#3885
·
phu0ngng
opened
Nov 17, 2026
Collaborator
·
13 tasks
·
·
2
[JAX] Fix EP=1 case in TE MoEBlock
#3884
·
tdophung
opened
Nov 17, 2026
Collaborator
·
7 of 13 tasks
·
·
4
[JAX] Control MeshResource at API call granularity for Dense APIs
#3883
·
tdophung
opened
Nov 17, 2026
Collaborator
·
8 of 13 tasks
·
·
2
[PyT][CP] A2A+Gemm/Gemm+A2A fusion
attention
#3881
·
sudhakarsingh27
opened
Nov 17, 2026
Member
·
13 tasks
·
·
4
[JAX] Fix encoder datasets on Python 3.14
org-contribution
#3880
·
aybchan
opened
Nov 17, 2026
Member
·
8 of 13 tasks
·
·
1
[Common] Use VR TMA override instruction for quantization kernels
#3879
·
Oleg-Goncharov
opened
Nov 17, 2026
Collaborator
·
4 of 13 tasks
·
·
[Docs] Name the parameters these five docstrings actually take
community-contribution
#3878
·
lxcxjxhx
opened
Nov 17, 2026
Contributor
·
·
8
Expose GDN state operations for chunkwise context parallelism
community-contribution
#3875
·
layalir
opened
Nov 16, 2026
Contributor
·
·
[PyTorch] Clone quantized tensor scaling buffers independently
community-contribution
#3874
·
ZenAlexa
opened
Nov 16, 2026
·
1 task done
·
·
4
[JAX] Enable cuDNN SiTU-GLU fusion in MoEBlock
#3873
·
pggPL
opened
Nov 16, 2026
Collaborator
·
7 of 13 tasks
·
·
9
[Common] Compute the capturable FusedAdam bias corrections once per block
community-contribution
#3872
·
xrennvidia
opened
Nov 15, 2026
Collaborator
·
7 of 13 tasks
·
·
1
[PyTorch] Run NVFP4 FC2 fprop and FC1 dgrad of the fused grouped MLP on CuTe DSL
community-contribution
#3871
·
xrennvidia
opened
Nov 15, 2026
Collaborator
·
8 of 13 tasks
·
·
1
[PyTorch] Multi-tensor NVFP4 2D master-weight cast
#3869
·
sraman-rgb
opened
Nov 13, 2026
Collaborator
·
13 tasks
·
·
2
[PyTorch][Common] Single-pass sigmoid router with aux-loss scores for…
#3868
·
sraman-rgb
opened
Nov 13, 2026
Collaborator
·
13 tasks
·
·
5
[Common][PyTorch] MLA q/kv RoPE + MXFP8 in attention's scale layout, …
#3867
·
sraman-rgb
opened
Nov 13, 2026
Collaborator
·
13 tasks
·
·
2
[Common][PyTorch] Performance Improvements for FP8BS emulation on Blackwell
#3866
·
denera
opened
Nov 13, 2026
Collaborator
·
9 of 13 tasks
·
·
4
[PyTorch] Add causal MXFP8 attention without future-token leakage through V scales
community-contribution
#3865
·
romitjain
opened
Nov 13, 2026
·
8 of 13 tasks
·
·
1
Previous
1
2
3
4
5
6
7
…
10
Next
You can’t perform that action at this time.