---
cited_by:
  - '[[thoughts/papers/qiu2026designqwen38nextarchitectureevaluation#11-qsa--qwen-sparse-attention|qwen4next §1.1]]'
  - '[[thoughts/inference/qwen3.8-serving#4-pr-reading-order-vllm-main-git-show-sha|qwen4next §4]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/warmup/kernel_warmup.py
merged: '2026-09-02'
number: 54513
source:
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/models/qwen4_exp/nvidia/ops/qsa_indexer.py
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/models/qwen4_exp/nvidia/ops/qsa.py
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/models/qwen4_exp/common/qsa_cache.py
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/models/qwen4_exp/nvidia/indexer_qsa.py
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/model_executor/warmup/qwen4_exp_qsa_warmup.py
  - https://github.com/vllm-project/vllm/blob/003e34341a824ab8e3bec4861e92b5e84180af8b/vllm/model_executor/warmup/kernel_warmup.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#54513 [Qwen3.8-Flash-Next] Separate prefill and decode paths for QSA indexer'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/54513
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/54513
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/54513.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Separate prefill and decode paths for QSA indexer · full dump: [[thoughts/inference/vllm-prs/54513|#54513]]

