Skip to content

Navigation Menu

Explore
By company size
By use case
By industry
View all solutions
Topics
- AI
- DevOps
- Security
- Software Development
- View all
Explore
- GitHub Sponsors
  Fund open source developers
- The ReadME Project
  GitHub community articles
Repositories
- Enterprise platform
  AI-powered developer platform
Available add-ons
Pricing

Search code, repositories, users, issues, pull requests...

Search

Clear

Search syntax tips

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Saved searches

Use saved searches to filter your results more quickly

Name

Query

To see all available qualifiers, see our documentation.

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

neuralmagic / vllm Public

forked from vllm-project/vllm

Notifications You must be signed in to change notification settings
Fork 1
Star 7

Code
Pull requests 22
Actions
Projects
Security
Insights

Additional navigation options

Code
Pull requests
Actions
Projects
Security
Insights

[torch.compile] Dynamic fp8 + rms_norm fusion#31

Open

ProExpertProg wants to merge 21 commits intomainneuralmagic/vllm:mainfrom luka/rms-norm-fusion-refactorneuralmagic/vllm:luka/rms-norm-fusion-refactorCopy head branch name to clipboard

+1,736-252

Conversation Commits21 (21)Checks Files changed

Commits

Commits on Dec 11, 2024

Refactor fusion patterns into class
ProExpertProg
committed
Improved comments, move utils to MultiOutputMatch
ProExpertProg
committed
Allow multiple epsilons by clearing pattern matcher cache
ProExpertProg
committed
Add graph as property of match, add comments, add utilities, extract ops to constants
ProExpertProg
committed
dynamic quant (fused ops in python)
ProExpertProg
committed
add fused-rms-quant-dyn-per-token branch

Varun Sundar Rabindranath
authored and
ProExpertProg
committed
Upgrade reduction
ProExpertProg
committed
Use new dynamic ops for fusion, tolerance has to be higher.
ProExpertProg
committed
In progress dynamic fusion debugging
ProExpertProg
committed
Nit comment
ProExpertProg
committed
Fix fusion and tests to use dynamic per-token
ProExpertProg
committed
Remove debug graph output
ProExpertProg
committed
PR comments
ProExpertProg
committed
Fix TPU test
ProExpertProg
committed
Abstract out quantization type to allow more quant types more easily
ProExpertProg
committed
Extract getitem replacement and meta value fixing into insert_fused_node
ProExpertProg
committed
- OpOverload typing
ProExpertProg
committed
PR comments: kernels
ProExpertProg
committed
PR comments: opcheck
ProExpertProg
committed

Commits on Dec 12, 2024

Fix dispatch utils for AMD
ProExpertProg
committed
PR comments:
ProExpertProg
committed

Footer

© 2025 GitHub, Inc.

Footer navigation

Terms
Privacy
Security
Status
Docs
Contact

You can’t perform that action at this time.