---
cited_by:
  - '[[thoughts/inference/speculative-drafting#3-vllm-status-at-a-glance|reading-map §3]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/config/vllm.py
merged: '2026-09-22'
number: 58065
source:
  - https://github.com/vllm-project/vllm/blob/0961bbae2894d574be790d219651824eb199318e/vllm/config/vllm.py
state: merged
tags:
  - vllm
  - vllm-pr
  - specdec
title: '#58065 [Spec Decode] Enable async scheduling for DFlash'
topic: specdec
url: https://github.com/vllm-project/vllm/pull/58065
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/58065
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/58065.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Adds `dflash` to the async-scheduling allowlist in `VllmConfig`, on both the explicit `--async-scheduling` path (previously a hard `ValueError`) and the default auto-enable path (previously force-disabled with a warning). The V2 model… · full dump: [[thoughts/inference/vllm-prs/58065|#58065]]

