p PatchFirst

CVE-2026-54234

CVE-2026-54234 · vLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection

This page contains facts from public vulnerability sources. Check the original advisory before updating.

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.

CVSS 7.5 · High · EPSS 0.6% · Fix available

Published
Updated

Stack: python

Affected products and versions

  • vllm

    Versions
    • < 0.24.0: Affected

    No fix published

  • vllm PyPI

    Versions
    • 0.17.1
    • 0.18.0
    • 0.18.1
    • 0.19.0
    • 0.19.1
    • 0.20.0
    • 0.20.1
    • 0.20.2
    • 0.21.0
    • 0.22.0
    • 0.22.1
    • 0.23.0
    Affected ranges
    • ECOSYSTEM: ≥ 0.17.1, < 0.24.0

    Update to 0.24.0

Original sources

Check your server →