Is This vulnerability exploited in the wild?

No. Junglewise Threat Intelligence has no report of This vulnerability being exploited in the wild.

How severe is This vulnerability?

High severity, with a CVSS score of 7.5, published 21 October 2026.

Which products and versions are affected?

vllm (PyPI) < 0.31.0, ECOSYSTEM:introduced=0; vLLM Project vllm before 0.31.0.

Is there a fix for This vulnerability?

Yes: fixed in 0.31.0. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

How is This vulnerability attacked?

It is attacked over the network.

Check whether you run vllm (PyPI) or vLLM Project vllm at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

vLLM remote denial of service through forged multimodal UUID cache desynchronization

Q: Which products and versions are affected?

vllm (PyPI) < 0.31.0, ECOSYSTEM:introduced=0; vLLM Project vllm before 0.31.0.

Q: Is there a fix for This vulnerability?

Yes: fixed in 0.31.0. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

Q: How is This vulnerability attacked?

It is attacked over the network.

Q: Am I affected?

Check whether you run vllm (PyPI) or vLLM Project vllm at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

Severity: high · CVSS 7.5 · Published 2026-10-21

vLLM is a large language model serving engine used to deploy AI models. An unauthenticated attacker can crash the entire vLLM serving instance by sending specially crafted requests with forged multimodal cache identifiers, causing the engine process to die and requiring manual restart. This results in immediate service outage affecting all users.

Technologies: vllm (PyPI), vLLM Project vllm. Vendors: PyPI, vLLM Project.

Quick answers

Is This vulnerability exploited in the wild?: No. Junglewise Threat Intelligence has no report of This vulnerability being exploited in the wild.
How severe is This vulnerability?: High severity, with a CVSS score of 7.5, published 21 October 2026.
Which products and versions are affected?: vllm (PyPI) < 0.31.0, ECOSYSTEM:introduced=0; vLLM Project vllm before 0.31.0.
Is there a fix for This vulnerability?: Yes: fixed in 0.31.0. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is This vulnerability attacked?: It is attacked over the network.
Am I affected?: Check whether you run vllm (PyPI) or vLLM Project vllm at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

Fixed in: 0.31.0

Technical details

vLLM uses client-supplied UUID values verbatim as cache identities for multimodal content without validation. An attacker exploits desynchronization between the P0 (frontend) and P1 (engine) processor caches—which have independent LRU eviction policies—by seeding a UUID, flooding the P0 cache with unvalidated requests to evict it, then reusing the UUID with different content. This causes a token-placeholder mismatch in the engine, triggering an uncaught ValueError that crashes the EngineCore process and terminates all API servers.

Affected products

vLLM Project vLLM before 0.31.0

Timeline

2026-10-21: disclosed

References

https://github.com/vllm-project/vllm/security/advisories/GHSA-p92p-rxj5-7p2x

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.