Master: Model Support & nvFP4 Quantization

Open
#5,288 0 comments 0 reactions 1 assignee View on GitHub

@jjsjann123 is already working on this.

Since Oct 7, 2025.

Assessment

This issue has not been assessed yet.

Description

MoE Inference nvFP4 Priority = P0
  • nvFP4 Linear and and Grouped Linear support for Routed Experts
  • Gating logic for routing plus data preparation for Grouped Linears. The gating logic is different for Llama 4, GPT-OSS, Deepseek R1, and Qwen3-Next
  • GPT OSS uses sliding window attention. According to the "Sliding Window + Causal" section in https://pytorch.org/blog/flexattention/, there's going to be a big penalty if we implement this inefficiently. For example, F.scaled_dot_product_attention is going to be ~3x slower than flexattention.

Subtasks:

Dominant language
C++
Stars
404
Forks
84
PR merge metrics
No merged PRs in 30d

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA/Fuser

All issues in NVIDIA/Fuser

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.