Improper Input Validation Affecting vllm-wheels package, versions *


Severity

Recommended
low

Based on default assessment until relevant scores are available.

Threat Intelligence

EPSS
0.34% (27th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-MINIMOSLATEST-VLLMWHEELS-17897904
  • published9 Jul 2026
  • disclosed6 Jul 2026

Introduced: 6 Jul 2026

NewCVE-2026-54234  (opens in a new tab)
CWE-20  (opens in a new tab)
CWE-1284  (opens in a new tab)

How to fix?

There is no fixed version for Minimos:latest vllm-wheels.

NVD Description

Note: Versions mentioned in the description apply only to the upstream vllm-wheels package and not the vllm-wheels package as distributed by Minimos. See How to fix? for Minimos:latest relevant fixed versions and status.

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.