Skip to content
Patch Next
Sign inAdd to your agent

← Actionable
Medium, CVSS 6.5amended

vLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertion in the engine core

CVE-2026-105753GHSA-ph3r-5jfg-f84f

What to do

Vendor fix available. Upgrade each affected package to the fixed version on its release line (vllm 0.28.0), redeploy, then verify what is actually installed.

This page has no inventory attached, so it cannot say whether you are affected. A vendor fix is not proof that your deployment is fixed.

Add to your agent

Affected packages

  • PyPI vllm

    Affected
    < 0.28.0
    Fixed in
    0.28.0

Sources

Severity and exploitation
Exploitation
No confirmed exploitation recorded. Patch Next marks exploitation only from CISA KEV.
CVSS
6.5 · CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
EPSS
  • CVE-2026-105753: 0.43% estimated exploitation activity in the next 30 days; percentile 37.04 (FIRST EPSS, 2026-11-03). This is a model estimate, not proof of exploitation or customer exposure.
Source description

As published by the source; long descriptions are shortened.

Affected - Ecosystem / package: pip / vllm - Affected versions: vLLM ≤ 0.25.1 (confirmed on 0.25.1, commit 752a3a504485). The lower bound predates 0.25.1; maintainers can confirm how far back the mirrored sender/receiver cache protocol reaches. Summary vLLM's default multimodal cache (mmprocessorcachetype="lru") mirrors state across two processes: the frontend (P0) holds only metadata (MultiModalProcessorSenderCache) while the engine core (P1) holds the real payload (MultiModalReceiverCache). The design invariant is that getandupdate() runs on P0 and P1 in lockstep for every request, so eviction order stays mirrored and P0 can answer "is this cached in P1?" without talking to P1. That invariant breaks when a request is rejected after P0 has rendered and hashed the multimodal input (populating the P0 cache) but before P1 receives the item — for example, an oversized chat prompt rejected on maxmodellen after rendering. P0 now believes the media is cached while P1 never got it. A later…

Dates and revisions
Published by the source
Last modified by the source
Source last checked
  1. Revision 3 · amended ·

    Amended from ghsa record GHSA-ph3r-5jfg-f84f

  2. Revision 2 · amended ·

    Amended from osv record GHSA-ph3r-5jfg-f84f

  3. Revision 1 · created ·

    Published from ghsa record GHSA-ph3r-5jfg-f84f

Agent brief

Preview the exact text before you copy it. Commands are read-only suggestions; review them before running anything.

# Patch Next advisory brief: CVE-2026-105753

Advisory: vLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertion in the engine core
IDs: CVE-2026-105753, GHSA-ph3r-5jfg-f84f
Record: https://patchnext.com/advisories/cve-2026-105753 (revision 3, source checked 26 Oct 2026, 13:55 UTC)
Inventory: none attached. This brief comes from a public page and does not say whether any deployment is affected.
Severity: Medium (CVSS 6.5)
Exploitation: no confirmed exploitation recorded. Patch Next confirms exploitation only from CISA KEV.
Vendor response: Vendor fix available; fixed versions by package: vllm 0.28.0. A vendor fix does not mean a deployment is fixed.

## Affected components
- PyPI vllm: affected < 0.28.0; fixed in 0.28.0

## Suggested inspection (read-only)
- `python -m pip show vllm`: prints the installed version. Run with the environment that serves production active.

## Supported action
Upgrade each affected component to a fixed version on its release line. For a transitive dependency, upgrade the parent package or add an override or constraint, then regenerate the lockfile. Redeploy every environment that ran an affected version.

## Required regression checks
- Run the test suite and a production build.
- Exercise the features that depend on vllm.
- Repeat the inspection and confirm that only fixed versions remain installed.

## Rollback considerations
- Keep the previous lockfile and deployment artifact. Rolling back reinstates the vulnerable version, so apply any documented mitigation while you fix forward.

## Unresolved questions
- Which deployments run an affected version?
- Exploitation status can change. Check again before lowering the priority.
- Mark the work resolved only after the fixed version is verified in production.

## Sources
- Verified: https://github.com/advisories/GHSA-ph3r-5jfg-f84f (GitHub-reviewed advisory GHSA-ph3r-5jfg-f84f lists affected ranges: PyPI vllm >=0 <0.28.0)
- Verified: https://osv.dev/vulnerability/GHSA-ph3r-5jfg-f84f (OSV record GHSA-ph3r-5jfg-f84f lists affected ranges: PyPI vllm >=0 <0.28.0 (fixed))

Commands are suggestions: review each one before running it.

Loading voting…