---
cited_by:
  - '[[thoughts/papers/kimiteam2026kimik3openfrontier#18-quant-formats|kimik3 §1.8]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/quantization/modelopt.py
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/fused_moe/experts/trtllm_nvfp4_moe.py
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/attention/mla_attention.py
merged: '2026-08-21'
number: 53132
source:
  - https://github.com/vllm-project/vllm/blob/f8e060271381e352a4eabec97f1839a32e41ec41/vllm/model_executor/layers/quantization/modelopt.py
  - https://github.com/vllm-project/vllm/blob/f8e060271381e352a4eabec97f1839a32e41ec41/vllm/model_executor/layers/fused_moe/experts/trtllm_nvfp4_moe.py
  - https://github.com/vllm-project/vllm/blob/f8e060271381e352a4eabec97f1839a32e41ec41/vllm/models/kimi_k3/nvidia/kda.py
  - https://github.com/vllm-project/vllm/blob/f8e060271381e352a4eabec97f1839a32e41ec41/vllm/model_executor/layers/attention/mla_attention.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#53132 Support kimi k3 nvfp4 checkpoint'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/53132
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/53132
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/53132.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
This PR supports running kimi k3 nvfp4 checkpoint `nvidia/Kimi-K3-NVFP4` · full dump: [[thoughts/inference/vllm-prs/53132|#53132]]

