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
/
Megatron-LM
Public
Notifications
You must be signed in to change notification settings
Fork
4.4k
Star
17.5k
Code
Issues
395
Pull requests
819
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
Make MTP / prefix cache stats persist for engine lifetime
- #4101
#4101
Merged
santhnm2
merged 27 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
santhnm2:mtp_prefix_caching_stats
santhnm2/Megatron-LM:mtp_prefix_caching_stats
Copy head branch name to clipboard
Jun 4, 2026
Conversation
Commits
27
(27)
Checks
Files changed
Merged
Make MTP / prefix cache stats persist for engine lifetime
#4101
santhnm2
merged 27 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
santhnm2:mtp_prefix_caching_stats
santhnm2/Megatron-LM:mtp_prefix_caching_stats
Copy head branch name to clipboard
Commits
Commits on Apr 1, 2026
Make MTP / prefix cache states persist for engine lifetime
Show description for 8f35b31
santhnm2
committed
8f35b31
View commit details
Copy full SHA for 8f35b31
Browse repository at this point
Commits on Apr 9, 2026
Merge with main
Show description for e16d0e2
santhnm2
committed
e16d0e2
View commit details
Copy full SHA for e16d0e2
Browse repository at this point
Track per-token acceptance rates
Show description for 917dc1b
santhnm2
committed
917dc1b
View commit details
Copy full SHA for 917dc1b
Browse repository at this point
Commits on May 20, 2026
Merge with main
Show description for 123cac4
santhnm2
committed
123cac4
View commit details
Copy full SHA for 123cac4
Browse repository at this point
Fix looping
Show description for 78f0ce9
santhnm2
committed
78f0ce9
View commit details
Copy full SHA for 78f0ce9
Browse repository at this point
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
529f04f
View commit details
Copy full SHA for 529f04f
Browse repository at this point
Remove unnecessary line
Show description for add45a4
santhnm2
committed
add45a4
View commit details
Copy full SHA for add45a4
Browse repository at this point
Commits on May 21, 2026
Replace num_mtp_heads with num_mtp_depths computed from model config
Show description for f68ccc3
santhnm2
and
claude
committed
f68ccc3
View commit details
Copy full SHA for f68ccc3
Browse repository at this point
Address reviewer comment
Show description for d9d00e1
santhnm2
committed
d9d00e1
View commit details
Copy full SHA for d9d00e1
Browse repository at this point
Linting
Show description for 1f5f73d
santhnm2
committed
1f5f73d
View commit details
Copy full SHA for 1f5f73d
Browse repository at this point
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
e9d4147
View commit details
Copy full SHA for e9d4147
Browse repository at this point
Commits on May 22, 2026
Return accepted_tokens=None for prefill-only batches in speculative decoding
Show description for a5cb845
santhnm2
and
claude
committed
a5cb845
View commit details
Copy full SHA for a5cb845
Browse repository at this point
Fix extra kwargs
Show description for 6439fa0
santhnm2
committed
6439fa0
View commit details
Copy full SHA for 6439fa0
Browse repository at this point
Linting
Show description for f08ed87
santhnm2
committed
f08ed87
View commit details
Copy full SHA for f08ed87
Browse repository at this point
Vectorize
Show description for 52e65c5
santhnm2
committed
52e65c5
View commit details
Copy full SHA for 52e65c5
Browse repository at this point
Linting
Show description for 777572b
santhnm2
committed
777572b
View commit details
Copy full SHA for 777572b
Browse repository at this point
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
6a87249
View commit details
Copy full SHA for 6a87249
Browse repository at this point
Support full model cuda graphs
Show description for edb3957
santhnm2
committed
edb3957
View commit details
Copy full SHA for edb3957
Browse repository at this point
Revert "Support full model cuda graphs"
Show description for eb4c646
santhnm2
committed
eb4c646
View commit details
Copy full SHA for eb4c646
Browse repository at this point
Commits on Jun 1, 2026
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
ea634f9
View commit details
Copy full SHA for ea634f9
Browse repository at this point
Commits on Jun 3, 2026
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
491f263
View commit details
Copy full SHA for 491f263
Browse repository at this point
Commits on Jun 4, 2026
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
3fcd581
View commit details
Copy full SHA for 3fcd581
Browse repository at this point
Comment fixes
Show description for 4f9ee47
santhnm2
committed
4f9ee47
View commit details
Copy full SHA for 4f9ee47
Browse repository at this point
Fix unit tests
Show description for 6431116
santhnm2
committed
6431116
View commit details
Copy full SHA for 6431116
Browse repository at this point
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
40409f0
View commit details
Copy full SHA for 40409f0
Browse repository at this point
Fix tests
Show description for d1dff10
santhnm2
committed
d1dff10
View commit details
Copy full SHA for d1dff10
Browse repository at this point
Merge branch 'main' into mtp_prefix_caching_stats
santhnm2
authored
41b6d99
View commit details
Copy full SHA for 41b6d99
Browse repository at this point
You can’t perform that action at this time.
