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
393
Pull requests
808
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
Perf fix
- #4996
#4996
Merged
shanmugamr1992
merged 18 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
shanmugamr1992:perf-fix
shanmugamr1992/Megatron-LM:perf-fix
Copy head branch name to clipboard
May 27, 2026
Conversation
Commits
18
(18)
Checks
Files changed
Merged
Perf fix
#4996
shanmugamr1992
merged 18 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
shanmugamr1992:perf-fix
shanmugamr1992/Megatron-LM:perf-fix
Copy head branch name to clipboard
Commits
Commits on May 14, 2026
test: add inference performance test harness for GPT 583M, hybrid 2B, GPT 16B
Show description for 1404dbe
shanmugamr1992
and
claude
committed
1404dbe
View commit details
Copy full SHA for 1404dbe
Browse repository at this point
Commits on May 15, 2026
Merge branch 'main' into perf-tests
shanmugamr1992
authored
b90fdcd
View commit details
Copy full SHA for b90fdcd
Browse repository at this point
Updated to fix errors
shanmugamr1992
committed
08e3c68
View commit details
Copy full SHA for 08e3c68
Browse repository at this point
Commits on May 18, 2026
Merge branch 'main' into perf-tests
shanmugamr1992
authored
1e73e1e
View commit details
Copy full SHA for 1e73e1e
Browse repository at this point
Commits on May 19, 2026
Making mr run only on merge in slurm clusters
shanmugamr1992
authored
cce0734
View commit details
Copy full SHA for cce0734
Browse repository at this point
Merge branch 'main' into perf-tests
shanmugamr1992
authored
ce2c8c9
View commit details
Copy full SHA for ce2c8c9
Browse repository at this point
chore: untrack run-performance-tests skill and slash command
Show description for b571da0
shanmugamr1992
and
claude
committed
b571da0
View commit details
Copy full SHA for b571da0
Browse repository at this point
Commits on May 21, 2026
perf-tests: consolidate 583M throughput coverage into perf harness; add +20% upper bound
Show description for d181830
shanmugamr1992
and
claude
committed
d181830
View commit details
Copy full SHA for d181830
Browse repository at this point
Merge branch 'main' into perf-tests
Show description for 2d8c350
shanmugamr1992
and
claude
committed
2d8c350
View commit details
Copy full SHA for 2d8c350
Browse repository at this point
perf-tests: use gsm8k real prompts for MoE/hybrid models (review feedback)
Show description for ad2817f
shanmugamr1992
and
claude
committed
ad2817f
View commit details
Copy full SHA for ad2817f
Browse repository at this point
perf-tests: bump gpt_16b_perf OSL 128 -> 2048 (reviewer feedback)
Show description for ee08ccd
shanmugamr1992
and
claude
committed
ee08ccd
View commit details
Copy full SHA for ee08ccd
Browse repository at this point
perf-tests: add hybrid_nanov3_3b_perf (TP=1 PP=1 EP=8, 8 GPUs)
Show description for a325a51
shanmugamr1992
and
claude
committed
a325a51
View commit details
Copy full SHA for a325a51
Browse repository at this point
perf-tests: enable chunked prefill for gpt_16b (subset of reviewer's flag list)
Show description for 2fab93f
shanmugamr1992
and
claude
committed
2fab93f
View commit details
Copy full SHA for 2fab93f
Browse repository at this point
Commits on May 22, 2026
perf-tests: gpt_16b OSL 2048 -> 256 (shorten CI wall time)
Show description for 6f219c4
shanmugamr1992
and
claude
committed
6f219c4
View commit details
Copy full SHA for 6f219c4
Browse repository at this point
perf-tests: add MIT attribution for vendored gsm8k prompt subset
Show description for e1b02c7
shanmugamr1992
and
claude
committed
e1b02c7
View commit details
Copy full SHA for e1b02c7
Browse repository at this point
Commits on May 26, 2026
perf-tests: drop p99_latency_ms from asserted METRICS (flaky with N=5)
Show description for 2fa6dca
shanmugamr1992
and
claude
committed
2fa6dca
View commit details
Copy full SHA for 2fa6dca
Browse repository at this point
Merge remote-tracking branch 'autoformatter-remote/main' into perf-fix
Show description for 685c23e
shanmugamr1992
committed
685c23e
View commit details
Copy full SHA for 685c23e
Browse repository at this point
perf-tests: restore --log-dir/--tee/--redirects in run_perf_test.sh
Show description for 59ba2fd
shanmugamr1992
and
claude
committed
59ba2fd
View commit details
Copy full SHA for 59ba2fd
Browse repository at this point
You can’t perform that action at this time.
