CVE-2026-114050 vllm-project CVE debrief

Who should care

Defenders responsible for the vLLM engine and its deployment should be aware of this vulnerability. This includes developers, system administrators, and security professionals who manage or use the vLLM engine in their applications or services.

Why it matters

CVE-2026-114050 is significant for defenders because it allows an unauthenticated client to cause a remote crash of the vLLM engine. This can lead to denial of service (DoS) attacks, impacting service availability and reliability. Defenders should verify the version of vLLM in use, apply patches, and monitor for exploitation attempts.

• Potential for remote denial of service (DoS) attacks against the vLLM engine.
• Need for verification of vLLM version and application of patches.
• Importance of monitoring for unusual request patterns that could indicate exploitation attempts.
• Potential impact on service availability and reliability.

Technical summary

The vLLM engine is vulnerable to a remote crash caused by an unauthenticated client. This is due to the use of user-supplied UUIDs as cache identities without proper synchronization or consistency validation between the frontend and engine processor caches. An attacker can exploit this by sending a series of requests with unique UUIDs and specific payloads to fill the cache and cause a crash. The vulnerability has been identified in the vLLM engine, and defenders should prioritize verifying the version of vLLM in use and applying the patch from version 0.31.0 or later. This involves checking the installed version of the vLLM package and updating it if necessary. Additionally, defenders should review and adjust

Defensive priority

Defenders should prioritize verifying the version of vLLM in use and applying the patch from version 0.31.0 or later. This involves checking the installed version of the vLLM package and updating it if necessary.

Recommended defensive actions

• Verify the version of vLLM in use and update to version 0.31.0 or later if necessary.
• Review and adjust cache eviction policies and synchronization mechanisms.
• Monitor for unusual patterns of requests that could indicate attempted exploitation.
• Check relevant monitoring, detection, and logs for exposed assets that need extra review.
• Track exceptions, retest remediated assets, and close the item only after evidence is documented.
• Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up.
• Plan vendor-supported updates or mitigations through normal change control where exposure is confirmed.

Evidence notes

The evidence provided includes details from the CVE record, the source item from osv_dev, and references from GitHub and NIST NVD. However, the corpus does not provide explicit information on the affected versions or exploitation attempts.

Sources and references

Methodology and review provenance

AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.