---
cited_by:
  - '[[thoughts/inference/kimi-k3-state#34-pd-dcp-kv-offload-59229-test-matrix|kimik3 §3.4]]'
  - '[[thoughts/inference/kimi-k3-prs#4-pr-reading-order|kimik3 §4]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/worker/gpu/model_runner.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/core/kv_cache_coordinator.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/worker/gpu/block_table.py
merged: '2026-08-18'
number: 50493
source:
  - https://github.com/vllm-project/vllm/blob/0db502c8d8a620f29dd3560e69a22f0a7ed3477c/vllm/v1/worker/gpu/model_runner.py
  - https://github.com/vllm-project/vllm/blob/0db502c8d8a620f29dd3560e69a22f0a7ed3477c/vllm/v1/core/kv_cache_coordinator.py
  - https://github.com/vllm-project/vllm/blob/0db502c8d8a620f29dd3560e69a22f0a7ed3477c/vllm/v1/worker/gpu/block_table.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#50493 [Kimi-K3] support DCP partial prefix cache hit'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/50493
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/50493
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/50493.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Builds on the Kimi-K3 DCP support from [#50484](https://github.com/aarnphm/aarnphm.github.io/issues/50484). · full dump: [[thoughts/inference/vllm-prs/50493|#50493]]

