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
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
/
TensorRT-LLM
Public
Notifications
You must be signed in to change notification settings
Fork
2.7k
Star
14.4k
Code
Issues
647
Pull requests
898
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
[None][refactor] Refine Skip Softmax follow-ups
- #15417
#15417
Merged
bobboli
merged 8 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
bobboli:bli/skip-softmax-followups-placeholder
bobboli/TensorRT-LLM:bli/skip-softmax-followups-placeholder
Copy head branch name to clipboard
Jul 7, 2026
Conversation
Commits
8
(8)
Checks
Files changed
Merged
[None][refactor] Refine Skip Softmax follow-ups
#15417
bobboli
merged 8 commits into
NVIDIA:main
NVIDIA/TensorRT-LLM:main
from
bobboli:bli/skip-softmax-followups-placeholder
bobboli/TensorRT-LLM:bli/skip-softmax-followups-placeholder
Copy head branch name to clipboard
Commits
Commits on Jul 6, 2026
Refine sparse attention backend plumbing
Show description for 96b9c86
bobboli
committed
96b9c86
View commit details
Copy full SHA for 96b9c86
Browse repository at this point
[None][fix] Keep hybrid KV routing for Skip Softmax
Show description for aec8393
bobboli
committed
aec8393
View commit details
Copy full SHA for aec8393
Browse repository at this point
[None][fix] Split Cosmos raw and normalized timesteps
Show description for 54afabf
bobboli
committed
54afabf
View commit details
Copy full SHA for 54afabf
Browse repository at this point
[None][fix] Update SAGE test sparse backend lookup
Show description for 006114e
bobboli
committed
006114e
View commit details
Copy full SHA for 006114e
Browse repository at this point
[None][fix] Lower DeepSeek V4 sparse params consistently
Show description for 00c31a9
bobboli
committed
00c31a9
View commit details
Copy full SHA for 00c31a9
Browse repository at this point
[None][fix] Use paged context FMHA with FP8 KV cache
Show description for 1e852d0
bobboli
committed
1e852d0
View commit details
Copy full SHA for 1e852d0
Browse repository at this point
[None][fix] Clarify FP8 context routing
Show description for a67650c
bobboli
committed
a67650c
View commit details
Copy full SHA for a67650c
Browse repository at this point
[None][test] Update Cosmos3 audio timestep inputs
Show description for 41f8c00
bobboli
committed
41f8c00
View commit details
Copy full SHA for 41f8c00
Browse repository at this point
You can’t perform that action at this time.