CVE-2026-93436
Medium
Elevated severity or exploit probability.
CVSS base
7.5
HIGH
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
EPSS — probability of exploitation (30 days)
0.8%
53.3th percentile
CISA KEV
Not listed
Weakness / dates
CWE-401
Published 2026-09-17 · modified 2026-09-22
Description
vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts.
References
- https://github.com/vllm-project/vllm
- https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/nixl/push_worker.py#L162-L181
- https://github.com/vllm-project/vllm/pull/55677
- https://www.vulncheck.com/advisories/vllm-through-0.29.0-memory-exhaustion-via-rejected-requests