---
cited_by:
  - '[[thoughts/inference/kv-transfer#5-reading-order|reading-map §5]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/attention/backends/flashinfer.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/attention/backends/utils.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/attention/backends/flash_attn.py
  - file:///Users/aarnphm/workspace/vllm/vllm/envs.py
  - file:///Users/aarnphm/workspace/vllm/vllm/distributed/kv_transfer/kv_connector/utils.py
merged: '2025-06-17'
number: 19280
source:
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/v1/attention/backends/flashinfer.py
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/v1/attention/backends/utils.py
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/v1/attention/backends/flash_attn.py
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/envs.py
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/distributed/kv_transfer/kv_connector/utils.py
  - https://github.com/vllm-project/vllm/blob/4c8f64faa74132580ecb5f5ed5184821b5fbb013/vllm/attention/backends/flashinfer.py
state: merged
tags:
  - vllm
  - vllm-pr
  - hetero-tp-kv-layout
title: '#19280 [V1][Kernel] Flashinfer HND KV cache layout'
topic: hetero-tp-kv-layout
url: https://github.com/vllm-project/vllm/pull/19280
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/19280
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/19280.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Follow up PR to again porting over functionality from V0 (ref · full dump: [[thoughts/inference/vllm-prs/19280|#19280]]

