Skip to main content
VulnerabilityPublished

CVE-2026-54234: vllm · Improper Input Validation

Affects vllmSource CVE Database V5

0
Compare

Not known to be exploited · CVSS 7.5 High : upgrade to 0.24.0 soon.

Data as of 18 Nov 2026, 20:25 UTC

CVSS 3.1 base score
7.5High
How severe the flaw is on the 0–10 CVSS scale.
EPSS
0.6% chance of exploitation in 30 days – higher than 50% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.
Patch
Available – fixed in 0.24.0
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H

Exploitation evidence

  1. Disclosed: evidence found.

    CVE published 6 Jul 2026

  2. Detectable: no evidence on record.

    No public nuclei template indexed

  3. Weaponized: no evidence on record.

    No public exploit code indexed in Metasploit or Exploit-DB

  4. Exploited in the wild: no evidence on record.

    Not in CISA KEV and no source reports exploitation

  5. Observed by OffSeq Mirage: no evidence on record.

    No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 18 Nov 2026.

What to do

Upgrade vllm and vllm-project/vllm to 0.24.0.

Affected software

ProductAffectedFixed in
vllm-projectvllmAffected
  • <0.24.0
Fixed in
  • 0.24.0
GitHubvllm-project/vllmpkg:github/vllm-project/vllmOther threatsAffected
  • <0.24.0
Fixed in
  • 0.24.0

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

Upgrade vLLM to version 0.24.0 or later, where this vulnerability is fixed. Until upgraded, restrict access to the public gRPC Generate and Abort endpoints to trusted clients to reduce exposure.

Description

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, a frontend-legal multi-request speculative decoding workload can cause the rejection sampler to produce a recovered token equal to the model vocabulary size boundary value, which is then converted to negative one when the engine selects the next live token for a request and is written back into the drafter's input ids; that out-of-vocabulary value is later consumed by the model's embedding and attention path and crashes the engine worker with a GPU device-side assertion. The same triggering request sequence is reachable through the public gRPC Generate and Abort endpoints, so a remote client that can send generation requests can crash the shared engine worker, aborting concurrent requests and causing a service-wide denial of service for other clients of the deployment until the worker is restarted. This issue is fixed in version 0.24.0.

References

Community reviews

Loading reviews…