Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🛡️
Not in CISA KEV
HIGH severity

CVE-2026-94623 Vllm-project

HIGHFix: vllm-project/vllm#51505

CVE-2026-94623 is a high-severity (CVSS 7.5) Reachable Assertion vulnerability. No vendor fix is recorded yet; mitigation options are listed below.

vLLM through 0.29.0 contains a denial of service vulnerability in the NIXL connector's prefix caching implementation that fails to properly validate block counts across multi-prompt…

Published
Sep 21, 2026
Updated
Sep 21, 2026
Affected
Patched
See advisory
Exploits
None indexed
Exploitation data as of Sep 21, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Description

vLLM through 0.29.0 contains a denial of service vulnerability in the NIXL connector's prefix caching implementation that fails to properly validate block counts across multi-prompt completion requests in prefill/decode disaggregated deployments. Attackers can trigger an assertion failure in NixlBaseConnectorWorker._apply_prefix_caching by submitting completion requests with multiple prompts of varying lengths, causing the decode worker to terminate and become unavailable until restarted.

Detection & mitigation playbook

Vulnerability
  1. Detect

    Identify every host running the affected component and compare the installed build against the fixed version below — for source-built or distro-packaged software the version string, not a lockfile, is the source of truth (`dpkg -l`, `rpm -q`, or the binary's own `--version`). Because exploitation targets the running service rather than your application code, O3 also flags the exploit behaviour from runtime telemetry and egress traffic.

  2. Remediation status

    No fixed release is recorded for CVE-2026-94623 yet. Track the upstream advisory, and apply the exposure-reduction steps below in the meantime.

  3. Mitigate without a patch

    Reduce reachability while you patch: restrict network access to the service to trusted sources, disable or gate the affected component where your deployment allows it, and add a WAF/IPS signature if one exists for CVE-2026-94623. O3's runtime protection blocks the reachable assertion flaw at execution, which holds the line on hosts you cannot patch immediately.

  4. How O3 protects you

    O3 detects and blocks CVE-2026-94623 exploitation at runtime with eBPF exploit-chain detection, plus L7 egress monitoring that catches a post-exploitation callback and cuts the attacker's outbound channel.

Tailored to CVE-2026-94623. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

vLLM through 0.29.0 contains a denial of service vulnerability in the NIXL connector's prefix caching implementation that fails to properly validate block counts across multi-prompt completion requests in prefill/decode disaggregated deployments. Attackers can trigger an assertion failure in NixlBaseConnectorWorker._apply_prefix_caching by submitting completion requests with multiple prompts of varying lengths, causing the decode worker to terminate and become unavailable until restarted.
O3 Security · Impact-Aware SCA

Is CVE-2026-94623 in your dependencies?

O3 Security finds CVE-2026-94623 across dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.

CVE-2026-94623: Vllm-project (High 7.5) | O3 Security