Skip to content
Navigation Menu
Sign in
Appearance settings
Platform
AI CODE CREATION
GitHub Copilot
Write better code with AI
GitHub Copilot app
Direct agents from issue to merge
MCP Registry
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
Code Quality
Enforce quality at merge
APPLICATION SECURITY
GitHub Advanced Security
Find and fix vulnerabilities
Code security
Secure your code as you build
Secret protection
Stop leaks before they start
EXPLORE
Why GitHub
Documentation
Blog
Changelog
Marketplace
View all features
Solutions
BY COMPANY SIZE
Enterprises
Small and medium teams
Startups
Nonprofits
BY USE CASE
App Modernization
DevSecOps
DevOps
CI/CD
View all use cases
BY INDUSTRY
Healthcare
Financial services
Manufacturing
Government
View all industries
View all solutions
Resources
EXPLORE BY TOPIC
AI
Software Development
DevOps
Security
View all topics
EXPLORE BY TYPE
Customer stories
Events & webinars
Ebooks & reports
Business insights
GitHub Skills
SUPPORT & SERVICES
Documentation
Customer support
Community forum
Trust center
Partners
View all resources
Open Source
COMMUNITY
GitHub Sponsors
Fund open source developers
PROGRAMS
Security Lab
Maintainer Community
Accelerator
GitHub Stars
Archive Program
REPOSITORIES
Topics
Trending
Collections
Enterprise
ENTERPRISE SOLUTIONS
Enterprise platform
AI-powered developer platform
AVAILABLE ADD-ONS
GitHub Advanced Security
Enterprise-grade security features
Copilot for Business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Pricing
Type
/
to 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 }}
Uh oh!
There was an error while loading.
Please reload this page
.
NVIDIA
/
TransformerEngine
Public
Notifications
You must be signed in to change notification settings
Fork
797
Star
3.5k
Code
Issues
157
Pull requests
160
Discussions
Actions
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Security and quality
Insights
[JAX] Expert Parallelism: JAX primitives + VJPs
- #3036
#3036
Merged
phu0ngng
merged 24 commits into
NVIDIA:main
NVIDIA/TransformerEngine:main
from
phu0ngng:phuong/ep-3-jax
phu0ngng/TransformerEngine:phuong/ep-3-jax
Copy head branch name to clipboard
Jun 27, 2026
Conversation
Commits
24
(24)
Checks
Files changed
Merged
[JAX] Expert Parallelism: JAX primitives + VJPs
#3036
phu0ngng
merged 24 commits into
NVIDIA:main
NVIDIA/TransformerEngine:main
from
phu0ngng:phuong/ep-3-jax
phu0ngng/TransformerEngine:phuong/ep-3-jax
Copy head branch name to clipboard
Commits
Commits on Jun 24, 2026
jax: add EP bindings on pointer-keyed cache with EpLayerConfig and bf16 max_token_dtype
Show description for 3ee9599
phu0ngng
committed
3ee9599
View commit details
Copy full SHA for 3ee9599
Browse repository at this point
jax/ep: drop topk_weights from ep_combine; caller must pre-multiply
Show description for fb3300f
phu0ngng
committed
fb3300f
View commit details
Copy full SHA for fb3300f
Browse repository at this point
tests/jax/ep: mask uninitialized recv_tokens tail in dispatch_vjp
Show description for fa907ca
phu0ngng
committed
fa907ca
View commit details
Copy full SHA for fa907ca
Browse repository at this point
examples/jax/ep: add ep_bench.py + run_ep_bench.sh
Show description for dedbf86
phu0ngng
committed
dedbf86
View commit details
Copy full SHA for dedbf86
Browse repository at this point
examples/jax/ep: ep_moe.py runs --iters fwd+bwd steps (default 3)
Show description for 453b399
phu0ngng
committed
453b399
View commit details
Copy full SHA for 453b399
Browse repository at this point
jax/ep: tighten sharding contract, drop helpers, route bwd through TE with_sharding_constraint
Show description for 4ee095c
phu0ngng
committed
4ee095c
View commit details
Copy full SHA for 4ee095c
Browse repository at this point
jax/ep: derive ep_size and num_ep_groups from active mesh in ep_bootstrap
Show description for 761d557
phu0ngng
committed
761d557
View commit details
Copy full SHA for 761d557
Browse repository at this point
examples/jax/ep: rename ep_handle to layer_cfg in ep_moe.py (matches EpLayerConfig type)
Show description for 82c9063
phu0ngng
committed
82c9063
View commit details
Copy full SHA for 82c9063
Browse repository at this point
jax/ep: add primitive docstrings and silence missing-kwoa false positives (lint 10.00)
Show description for a314a45
phu0ngng
committed
a314a45
View commit details
Copy full SHA for a314a45
Browse repository at this point
jax/ep: apply black formatting (pre-commit hook output)
Show description for c0b280d
phu0ngng
committed
c0b280d
View commit details
Copy full SHA for c0b280d
Browse repository at this point
build_tools/jax: gate NCCL EP on NVTE_BUILD_WITH_NCCL_EP (default on); define NVTE_WITH_NCCL_EP
Show description for 348cc36
phu0ngng
committed
348cc36
View commit details
Copy full SHA for 348cc36
Browse repository at this point
jax/ep: collapse 5 FFI attr structs into single EpConfig
Show description for bf811b8
phu0ngng
committed
bf811b8
View commit details
Copy full SHA for bf811b8
Browse repository at this point
jax/ep: dedup _ep_outer_axis, normalize _ep_spec_ok, unify outer_abstract, drop dead helpers
Show description for c030618
phu0ngng
committed
c030618
View commit details
Copy full SHA for c030618
Browse repository at this point
jax/ep: apply clang-format and silence pylint unused-arg in lowering
Show description for e41f588
phu0ngng
committed
e41f588
View commit details
Copy full SHA for e41f588
Browse repository at this point
qa: wire NCCL EP tests into L1 (multi-process unittest) and L0 (ep_moe example) jax distributed suites
Show description for 572889c
phu0ngng
committed
572889c
View commit details
Copy full SHA for 572889c
Browse repository at this point
jax/ep: resolve test path from launcher SCRIPT_DIR and export LD_LIBRARY_PATH for libnccl_ep.so
Show description for 943a970
phu0ngng
committed
943a970
View commit details
Copy full SHA for 943a970
Browse repository at this point
[JAX] Generalize EP MoE example to N stacked layers via --num-layers
Show description for 8818438
phu0ngng
committed
8818438
View commit details
Copy full SHA for 8818438
Browse repository at this point
Unify NCCL EP build flag: rename to NVTE_WITH_NCCL_EP, share arch check via nccl_ep_enabled()
Show description for a3378db
phu0ngng
committed
a3378db
View commit details
Copy full SHA for a3378db
Browse repository at this point
Add arg descriptions to ep_bootstrap docstring
Show description for 06440a6
phu0ngng
committed
06440a6
View commit details
Copy full SHA for 06440a6
Browse repository at this point
[pre-commit.ci] auto fixes from pre-commit.com hooks
Show description for ebf5d96
pre-commit-ci[bot]
authored and
phu0ngng
committed
ebf5d96
View commit details
Copy full SHA for ebf5d96
Browse repository at this point
L0_jax_unittest: exclude multi-process EP test on single-node runner
Show description for 9df769a
phu0ngng
committed
9df769a
View commit details
Copy full SHA for 9df769a
Browse repository at this point
Commits on Jun 25, 2026
Guard against None out_partition_spec in EP combine/dispatch-bwd partition methods
Show description for db8d43b
phu0ngng
committed
db8d43b
View commit details
Copy full SHA for db8d43b
Browse repository at this point
Commits on Jun 26, 2026
Fix test_two_layer_dispatch_no_handle_aliasing: add A->B data edge
Show description for 0a29355
phu0ngng
committed
0a29355
View commit details
Copy full SHA for 0a29355
Browse repository at this point
Commits on Jun 27, 2026
Merge branch 'main' into phuong/ep-3-jax
phu0ngng
authored
9187d81
View commit details
Copy full SHA for 9187d81
Browse repository at this point
You can’t perform that action at this time.