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
/
Megatron-LM
Public
Notifications
You must be signed in to change notification settings
Fork
4.6k
Star
18.1k
Code
Issues
454
Pull requests
1.1k
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
MInf: Add hooks to preprocess and stage
- #7015
#7015
Merged
tdene
merged 25 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
tdene:tde/ledger_capture
tdene/Megatron-LM:tde/ledger_capture
Copy head branch name to clipboard
Sep 19, 2026
Conversation
Commits
25
(25)
Checks
Files changed
Merged
MInf: Add hooks to preprocess and stage
#7015
tdene
merged 25 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
tdene:tde/ledger_capture
tdene/Megatron-LM:tde/ledger_capture
Copy head branch name to clipboard
Commits
Commits on Sep 3, 2026
Add a payload offload mode with a staging protocol
Show description for b48ab09
tdene
authored and
lauradang
committed
b48ab09
View commit details
Copy full SHA for b48ab09
Browse repository at this point
Accept `required_prefix_token_ids` on the endpoint
Show description for 789f746
tdene
authored and
lauradang
committed
789f746
View commit details
Copy full SHA for 789f746
Browse repository at this point
Commits on Sep 9, 2026
feat(inference): add canonical token capture hooks
Show description for 32c27fa
lauradang
committed
32c27fa
View commit details
Copy full SHA for 32c27fa
Browse repository at this point
fix(inference): define prefix replacement boundary helper
Show description for 086b755
lauradang
committed
086b755
View commit details
Copy full SHA for 086b755
Browse repository at this point
fix(inference): sort capture hook imports
Show description for fc3b3bd
lauradang
committed
fc3b3bd
View commit details
Copy full SHA for fc3b3bd
Browse repository at this point
Commits on Sep 10, 2026
fix(inference): serialize merged finished requests
Show description for eb245dd
lauradang
committed
eb245dd
View commit details
Copy full SHA for eb245dd
Browse repository at this point
fix(inference): satisfy prefix capture control flow lint
Show description for cd19bd6
lauradang
committed
cd19bd6
View commit details
Copy full SHA for cd19bd6
Browse repository at this point
fix(inference): forward opaque request metadata
Show description for fa751da
lauradang
committed
fa751da
View commit details
Copy full SHA for fa751da
Browse repository at this point
fix(inference): preserve payload response contracts
Show description for db79fa5
lauradang
committed
db79fa5
View commit details
Copy full SHA for db79fa5
Browse repository at this point
Commits on Sep 14, 2026
feat(inference): forward the prefix splice boundary to the engine prompt preparer
Show description for 26e82ea
lauradang
and
claude
committed
26e82ea
View commit details
Copy full SHA for 26e82ea
Browse repository at this point
Merge branch 'main' into tde/ledger_capture
lauradang
authored
4cdad25
View commit details
Copy full SHA for 4cdad25
Browse repository at this point
Guard einops import in multimodal utils
Show description for e7f245e
lauradang
and
claude
committed
e7f245e
View commit details
Copy full SHA for e7f245e
Browse repository at this point
Commits on Sep 15, 2026
Merge branch 'main' into tde/ledger_capture
Show description for ce2c8ab
lauradang
and
claude
committed
ce2c8ab
View commit details
Copy full SHA for ce2c8ab
Browse repository at this point
Commits on Sep 17, 2026
Rename the opaque request_metadata dict to offload_params
Show description for 0116fab
lauradang
and
claude
committed
0116fab
View commit details
Copy full SHA for 0116fab
Browse repository at this point
Let the prompt preparer replace prefix tokens with its own splicer
Show description for f050d85
lauradang
and
claude
committed
f050d85
View commit details
Copy full SHA for f050d85
Browse repository at this point
Update unit tests for the main merge of the offload wire format
Show description for cdaf623
lauradang
and
claude
committed
cdaf623
View commit details
Copy full SHA for cdaf623
Browse repository at this point
Harden prompt preparation failures in the dynamic engine
Show description for 8384f1b
lauradang
and
claude
committed
8384f1b
View commit details
Copy full SHA for 8384f1b
Browse repository at this point
Carry offload_params in its own SUBMIT_REQUEST frame
Show description for 2e00d27
lauradang
and
claude
committed
2e00d27
View commit details
Copy full SHA for 2e00d27
Browse repository at this point
Drop prompt tensors from offloaded replies and tidy the offload wire contract
Show description for 485bd65
lauradang
and
claude
committed
485bd65
View commit details
Copy full SHA for 485bd65
Browse repository at this point
Fix stale variable name in test_payload_offload_mode
Show description for 01f1247
lauradang
and
claude
committed
01f1247
View commit details
Copy full SHA for 01f1247
Browse repository at this point
Surface stage metadata on /v1/completions and reserve underscore keys in offload_params
Show description for 2800313
lauradang
and
claude
committed
2800313
View commit details
Copy full SHA for 2800313
Browse repository at this point
fix(inference): type prompt preparation results
Show description for 1310225
lauradang
committed
1310225
View commit details
Copy full SHA for 1310225
Browse repository at this point
Commits on Sep 18, 2026
Merge branch 'main' into tde/ledger_capture
Show description for dd5439b
lauradang
and
claude
committed
dd5439b
View commit details
Copy full SHA for dd5439b
Browse repository at this point
Fix dynamic engine tests broken by the main merge
Show description for 6dc0789
lauradang
and
claude
committed
6dc0789
View commit details
Copy full SHA for 6dc0789
Browse repository at this point
Merge branch 'main' into tde/ledger_capture
Show description for d446a91
lauradang
and
claude
committed
d446a91
View commit details
Copy full SHA for d446a91
Browse repository at this point
You can’t perform that action at this time.
