---
cited_by:
  - '[[thoughts/inference/kimi-k3#5-sharp-edges-each-fix-shows-an-invariant|kimik3 §5]]'
date: '2026-11-11'
kind: pr
local: []
merged: '2026-09-08'
number: 53379
source:
  - https://github.com/vllm-project/vllm/blob/f998862d46e7af54467e10156f02cc51a67e268b/vllm/models/kimi_k3/nvidia/model.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#53379 [Bugfix] Fix Kimi K3 loading with interleaved weight streams'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/53379
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/53379
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/53379.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Fix Kimi K3 weight loading when a streaming loader yields non-contiguous top-level module prefixes. · full dump: [[thoughts/inference/vllm-prs/53379|#53379]]

