---
cited_by:
  - '[[thoughts/inference/architectures#5-four-things-called-replay|README §5]]'
  - '[[thoughts/inference/deepseek-v4.1-state#34-not-replayssm|dsv41 §3.4]]'
  - '[[thoughts/inference/kimi-k3-state#33-spec-decode-with-recurrent-state|kimik3 §3.3]]'
  - '[[thoughts/inference/replayssm-serving#5-pr-reading-order|replayssm §5]]'
  - '[[thoughts/inference/replayssm#unverified--gaps|replayssm § Unverified / gaps]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/mamba/ops/ssu_dispatch.py
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/mamba/mamba_mixer2.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/attention/backends/mamba_attn.py
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/layers/mamba/mamba_utils.py
  - file:///Users/aarnphm/workspace/vllm/vllm/v1/worker/gpu/model_runner.py
  - file:///Users/aarnphm/workspace/vllm/vllm/config/vllm.py
  - file:///Users/aarnphm/workspace/vllm/vllm/model_executor/models/nemotron_h.py
merged: '2026-09-01'
number: 52506
source:
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/model_executor/layers/mamba/ops/ssu_dispatch.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/model_executor/layers/mamba/mamba_mixer2.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/model_executor/warmup/replayssm_warmup.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/v1/attention/backends/mamba_attn.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/model_executor/layers/mamba/mamba_utils.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/v1/worker/gpu/model_runner.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/config/vllm.py
  - https://github.com/vllm-project/vllm/blob/adebc41b7e9f1085d3f73434e23beb76883b9eb4/vllm/model_executor/models/nemotron_h.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#52506 [Mamba] Add FlashInfer ReplaySSM backend'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/52506
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/52506
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/52506.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
This PR adds FlashInfer as an optional ReplaySSM backend for Mamba2. It calls FlashInfer’s native `checkpointing_ssu` API, shares replay trackers per KV-cache group, and autotunes a maximum-batch decode through the existing FlashInfer… · full dump: [[thoughts/inference/vllm-prs/52506|#52506]]

