CVE-2026-94622: Uncaught Exception in vllm-project vllm
vLLM versions up to 0.29.0 have a denial of service vulnerability caused by an uncaught KeyError in the NIXL connector's metadata handling. This occurs when requests contain incomplete kv_transfer_params dictionary entries, leading the decode engine to terminate and causing all routed requests to fail until manually restarted.
AI Analysis
Technical Summary
CVE-2026-94622 affects vLLM versions through 0.29.0. The vulnerability arises in the NIXL connector's metadata handling for prefill/decode disaggregated deployments. Specifically, if an attacker sends requests with incomplete kv_transfer_params dictionary entries, an uncaught KeyError is triggered in the EngineCore scheduling component. This exception causes the decode engine to terminate unexpectedly, resulting in a denial of service where all routed requests fail until the service is manually restarted.
Potential Impact
Successful exploitation causes a denial of service by terminating the decode engine, disrupting all routed requests until manual intervention restarts the service. There is no indication of privilege escalation, data disclosure, or code execution. The impact is limited to service availability.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, monitor for malformed requests with incomplete kv_transfer_params and consider implementing input validation or request filtering to prevent triggering the uncaught exception. Manual restart of the decode engine is required after an incident.
CVE-2026-94622: Uncaught Exception in vllm-project vllm
Description
vLLM versions up to 0.29.0 have a denial of service vulnerability caused by an uncaught KeyError in the NIXL connector's metadata handling. This occurs when requests contain incomplete kv_transfer_params dictionary entries, leading the decode engine to terminate and causing all routed requests to fail until manually restarted.
CVSS v4.0
Score 8.7high
Affected software
vllm-project
vllm
pkg:github/vllm-project/vllmRun on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
CVE-2026-94622 affects vLLM versions through 0.29.0. The vulnerability arises in the NIXL connector's metadata handling for prefill/decode disaggregated deployments. Specifically, if an attacker sends requests with incomplete kv_transfer_params dictionary entries, an uncaught KeyError is triggered in the EngineCore scheduling component. This exception causes the decode engine to terminate unexpectedly, resulting in a denial of service where all routed requests fail until the service is manually restarted.
Potential Impact
Successful exploitation causes a denial of service by terminating the decode engine, disrupting all routed requests until manual intervention restarts the service. There is no indication of privilege escalation, data disclosure, or code execution. The impact is limited to service availability.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, monitor for malformed requests with incomplete kv_transfer_params and consider implementing input validation or request filtering to prevent triggering the uncaught exception. Manual restart of the decode engine is required after an incident.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-09-21T21:42:24.373Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6ab1acdf55bf5e2cf58d35c0
Added to database: 09/21/2026, 22:17:03 UTC
Last enriched: 09/21/2026, 22:31:43 UTC
Last updated: 09/21/2026, 22:50:45 UTC
Views: 6
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.