---
cited_by:
  - '[[thoughts/inference/replayssm-serving#5-pr-reading-order|replayssm §5]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/attention/backends/flashinfer.py
  - file:///Users/aarnphm/workspace/vllm/vllm/utils/flashinfer.py
merged: '2026-08-11'
number: 49718
source:
  - https://github.com/vllm-project/vllm/blob/1d2d83a07fd3180068917a031c28dcc83141d0be/vllm/v1/attention/backends/flashinfer.py
  - https://github.com/vllm-project/vllm/blob/1d2d83a07fd3180068917a031c28dcc83141d0be/vllm/utils/flashinfer.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#49718 [Attention] Add FlashInfer XQA decode support on SM12x'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/49718
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/49718
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/49718.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Enable FlashInfer XQA decode on SM120/SM121 through FlashInfer’s dedicated XQA API. · full dump: [[thoughts/inference/vllm-prs/49718|#49718]]

