CVE-2026-114055: Uncontrolled Resource Consumption in vllm-project vllm
Description
vLLM versions from 0.7.3 up to but not including 0.31.0 contain an uncontrolled resource consumption vulnerability. Remote clients can manipulate multimodal processor parameters to exhaust server memory, causing termination of the shared server process. This affects processors such as Isaac, InternVL, and H2OVL. The vulnerability has a high severity with a CVSS score of 8.7.
CVSS v4.0
Score 8.7high
Affected software
vllm-project
vllm
Run 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-2026-114055 describes an uncontrolled resource consumption vulnerability in vLLM versions >=0.7.3 and <0.31.0. Remote attackers can supply per-request mm_processor_kwargs to control parameters like image size, patch count, crop count, or frame count in multimodal processors (Isaac, InternVL, H2OVL, and others). By setting these parameters to large values, attackers can exhaust memory resources on the shared server, leading to termination of the server process.
Potential Impact
Successful exploitation allows unauthenticated remote clients to cause denial of service by exhausting server memory resources, resulting in termination of the shared server process. This can disrupt availability of the vLLM service.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, restrict access to the vLLM service to trusted clients and monitor for unusual parameter values in requests to mitigate potential exploitation.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-10-20T21:42:27.669Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6ad80c012cdf04f6568caa1a
Added to database: 10/21/2026, 00:49:05 UTC
Last enriched: 10/21/2026, 01:34:09 UTC
Last updated: 10/21/2026, 02:28:26 UTC
Views: 2
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.