Skip to content
Patch Next
Sign inAdd to your agent

← Actionable
Medium, CVSS 6.5amended

vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters

CVE-2026-44223GHSA-83vm-p52w-f9pw

What to do

Vendor fix available. Upgrade each affected package to the fixed version on its release line (vllm 0.20.0), 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
    >= 0.18.0, < 0.20.0
    Fixed in
    0.20.0

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-2026-44223: 0.37% estimated exploitation activity in the next 30 days; percentile 30.60 (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.

Summary The extracthiddenstates speculative decoding proposer in vLLM returns a tensor with an incorrect shape after the first decode step, causing a RuntimeError that crashes the EngineCore process. The crash is triggered when any request in the batch uses sampling penalty parameters (repetitionpenalty, frequencypenalty, or presencepenalty). A single request with a penalty parameter (e.g., "repetitionpenalty": 1.1) is sufficient to crash the server. The crash is deterministic and immediate — no concurrency, race condition, or special workload is required. Details In vLLM v0.17.0, the extracthiddenstates proposer's propose() method returned sampledtokenids.unsqueeze(-1), producing a tensor of shape (batchsize, 1). In PR #37013 (first released in v0.18.0), the KV connector interface was refactored out of propose(). The return type changed from tuple[Tensor, KVConnectorOutput | None] to Tensor, and the .unsqueeze(-1) call was removed along with the KV connector output: The refactor…

Dates and revisions
Published by the source
Last modified by the source
Source last checked
  1. Revision 2 · amended ·

    Amended from ghsa record GHSA-83vm-p52w-f9pw

  2. Revision 1 · created ·

    Published from osv record GHSA-83vm-p52w-f9pw

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-2026-44223

Advisory: vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parameters
IDs: CVE-2026-44223, GHSA-83vm-p52w-f9pw
Record: https://patchnext.com/advisories/cve-2026-44223 (revision 2, source checked 26 Oct 2026, 13:44 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 0.20.0. A vendor fix does not mean a deployment is fixed.

## Affected components
- PyPI vllm: affected >= 0.18.0, < 0.20.0; fixed in 0.20.0

## 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://osv.dev/vulnerability/GHSA-83vm-p52w-f9pw (OSV record GHSA-83vm-p52w-f9pw lists affected ranges: PyPI vllm >=0.18.0 <0.20.0 (fixed))
- Verified: https://github.com/advisories/GHSA-83vm-p52w-f9pw (GitHub-reviewed advisory GHSA-83vm-p52w-f9pw lists affected ranges: PyPI vllm >=0.18.0 <0.20.0)

Commands are suggestions: review each one before running it.

Loading voting…