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
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
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
/
TensorRT-LLM
Public
Notifications
You must be signed in to change notification settings
Fork
2.8k
Star
14.7k
Code
Issues
606
Pull requests
899
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
[https://nvbugs/6705472][feat] FA4 variable-length cross-attention
- #18973
#18973
Draft
o-stoner
wants to merge 10 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
o-stoner:user/o-stoner/visual-gen-variable-len-attn
o-stoner/TensorRT-LLM:user/o-stoner/visual-gen-variable-len-attn
Copy head branch name to clipboard
Conversation
Commits
10
(10)
Checks
Files changed
Draft
[https://nvbugs/6705472][feat] FA4 variable-length cross-attention
#18973
o-stoner
wants to merge 10 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
o-stoner:user/o-stoner/visual-gen-variable-len-attn
o-stoner/TensorRT-LLM:user/o-stoner/visual-gen-variable-len-attn
Copy head branch name to clipboard
Commits
Commits on Sep 8, 2026
add FA4 cu_seqlens varlen cross-attention support
Show description for 171d8d9
o-stoner
committed
171d8d9
View commit details
Copy full SHA for 171d8d9
Browse repository at this point
trim varlen attention tests
Show description for 7c6b417
o-stoner
committed
7c6b417
View commit details
Copy full SHA for 7c6b417
Browse repository at this point
Commits on Sep 9, 2026
avoid forced host-device sync in pack_ragged_kv, fix varlen cross-attn test flakiness
Show description for 2cc7e14
o-stoner
committed
2cc7e14
View commit details
Copy full SHA for 2cc7e14
Browse repository at this point
padded q, varlen k/v for better perf
Show description for 4813317
o-stoner
committed
4813317
View commit details
Copy full SHA for 4813317
Browse repository at this point
speed up pack_ragged_kv: host-side cumsum, cache cu_seqlens_kv by kv_lens
Show description for 5120115
o-stoner
committed
5120115
View commit details
Copy full SHA for 5120115
Browse repository at this point
add production-scale and cache correctness tests for varlen cross-attention
Show description for 60bb3e8
o-stoner
committed
60bb3e8
View commit details
Copy full SHA for 60bb3e8
Browse repository at this point
Commits on Sep 10, 2026
drop public varlen_cfg knob, cache cu_seqlens_kv/max_seqlen_kv
Show description for fdaac48
o-stoner
committed
fdaac48
View commit details
Copy full SHA for fdaac48
Browse repository at this point
Commits on Sep 11, 2026
add varlen attention test to CI
Show description for ffedc89
o-stoner
committed
ffedc89
View commit details
Copy full SHA for ffedc89
Browse repository at this point
rename TestPackRaggedKvCache to TestPackRaggedKvMetadataCache
Show description for 98f85b7
o-stoner
committed
98f85b7
View commit details
Copy full SHA for 98f85b7
Browse repository at this point
Merge branch 'main' into user/o-stoner/visual-gen-variable-len-attn
Show description for 476ae55
o-stoner
authored
476ae55
View commit details
Copy full SHA for 476ae55
Browse repository at this point
You can’t perform that action at this time.
