Skip to content

Fix QMoE blockwise quantization support for TRT-RTX execution provider #1926

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
anujj wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Fix QMoE blockwise quantization support for TRT-RTX execution provider #1926

anujj wants to merge 4 commits into from
+208 −61

Conversation

@anujj
Copy link
Contributor

@anujj anujj commented Dec 19, 2025

Add QMoE and BF16 support for TRT-RTX execution provider

  • Enable blockwise quantization for TRT-RTX/NvTensorRtRtx EPs
  • Add gpt_oss_swiglu_fusion option for separate gate/up weights
  • Add int4_qdq_block_size for MatMul quantization block size
  • Add BF16 precision support for TRT-RTX
  • Keep padding in QMoE weights for proper alignment

@anujj anujj marked this pull request as draft December 19, 2025 13:35
@anujj
Copy link
Contributor Author

anujj commented Jan 6, 2026

@kunal-vaishnavi @baijumeswani for review

@anujj anujj marked this pull request as ready for review January 6, 2026 08:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@anujj