---
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/kv_cache_interface.py
  - file:///Users/aarnphm/workspace/vllm/vllm/distributed/kv_transfer/kv_connector/v1/nixl/worker.py
  - file:///Users/aarnphm/workspace/vllm/vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py
  - file:///Users/aarnphm/workspace/vllm/vllm/distributed/kv_transfer/kv_connector/v1/nixl/utils.py
number: 43151
source:
  - https://github.com/vllm-project/vllm/blob/03eca0d39b43d3dc8be4c13cf982b6218897604d/vllm/v1/kv_cache_interface.py
  - https://github.com/vllm-project/vllm/blob/03eca0d39b43d3dc8be4c13cf982b6218897604d/vllm/distributed/kv_transfer/kv_connector/v1/nixl/worker.py
  - https://github.com/vllm-project/vllm/blob/03eca0d39b43d3dc8be4c13cf982b6218897604d/vllm/distributed/kv_transfer/kv_connector/v1/nixl/tp_mapping.py
  - https://github.com/vllm-project/vllm/blob/03eca0d39b43d3dc8be4c13cf982b6218897604d/vllm/distributed/kv_transfer/kv_connector/v1/nixl/utils.py
state: closed
tags:
  - vllm
  - vllm-pr
  - hetero-tp-kv-layout
title: '#43151 [NIXL][2/N] Cache TP slicing and mapping redesign'
topic: hetero-tp-kv-layout
url: https://github.com/vllm-project/vllm/pull/43151
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/43151
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/43151.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
\[2/N\] Cache TP slicing and mapping redesign · full dump: [[thoughts/inference/vllm-prs/43151|#43151]]

