---
cited_by:
  - '[[thoughts/inference/architectures#3-what-changed-in-vllms-kv-cache-manager-to-hold-these|README §3]]'
  - '[[thoughts/inference/kimi-k3#16-speculative-decoding-mtp-dspark-eagle3dflash|kimik3 §1.6]]'
  - '[[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: []
merged: '2026-10-08'
number: 59733
source:
  - https://github.com/vllm-project/vllm/blob/1fbbfa165709c490dae48884eebdfbf6a1aef330/vllm/models/kimi_k3/nvidia/mla.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#59733 [Bugfix][Kimi-K3] Declare max_tp_shards on the DSpark MLA KV cache spec'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/59733
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/59733
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/59733.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Fixes a Kimi-K3 + DSpark perf regression from [#57652](https://github.com/aarnphm/aarnphm.github.io/issues/57652): the K3 DSpark draft’s `MLAAttentionSpec` does not declare `max_tp_shards`, so its MLA layers stop merging with the target’s and the KV cache splits from 4 to 20 groups. One-line fix:… · full dump: [[thoughts/inference/vllm-prs/59733|#59733]]

