CVE-2026-54234
vLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection
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.
| CWE | CWE-20 CWE-1284 |
| Vendor | vllm-project |
| Product | vllm |
| Published | Jul 6, 2026 |
| Last Updated | Jul 7, 2026 |
Get instant alerts for vllm-project vllm
Be the first to know when new high vulnerabilities affecting vllm-project vllm are published — delivered to Slack, Telegram or Discord.
CVSS v3 Breakdown
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H