CVE-2026-114051: Uncaught Exception in vllm-project vllm
Description
vLLM versions before 0.31.0 have a vulnerability where the min_tokens <= max_tokens check is skipped if max_tokens is null in /inference/v1/generate requests. This causes an uncaught exception that hangs all subsequent generation requests until the service is manually restarted. The vulnerability can be triggered by unauthenticated remote attackers sending a single request with an oversized min_tokens value.
CVSS v4.0
Score 7.1high
Affected software
vllm-project
vllm
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vulnerability in vLLM prior to version 0.31.0 involves skipping the validation check that min_tokens must be less than or equal to max_tokens when max_tokens is null in /inference/v1/generate API requests. This leads to an uncaught VLLMValidationError in the EngineCore input thread, causing it to die. Although the /health endpoint continues to return a 200 status, the service becomes unresponsive to generation requests until manually restarted. The flaw allows unauthenticated remote attackers to hang the service by sending a single malicious request with an oversized min_tokens parameter.
Potential Impact
An unauthenticated remote attacker can cause a denial of service by sending one specially crafted request that causes the input thread to crash, hanging all subsequent generation requests. The service appears healthy via the /health endpoint but is effectively non-functional until manually restarted. This results in a high-impact denial of service condition.
Mitigation Recommendations
A fix is available in vLLM version 0.31.0 and later. Users should upgrade to version 0.31.0 or newer to resolve this issue. Until upgraded, manual restarts are required to recover from the hang caused by this vulnerability.
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: 6ad80bff2cdf04f6568caa14
Added to database: 10/21/2026, 00:49:03 UTC
Last enriched: 10/21/2026, 01:34:20 UTC
Last updated: 10/21/2026, 01:34:20 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.