Bug 2546627 (CVE-2026-105922)

Summary: CVE-2026-105922 vllm: vLLM: Denial of Service via crafted inference request with prompt embeddings
Product: [Other] Security Response Reporter: OSIDB Bzimport <bzimport>
Component: vulnerabilityAssignee: Product Security DevOps Team <prodsec-dev>
Status: NEW --- QA Contact:
Severity: high Docs Contact:
Priority: high    
Version: unspecifiedCC: alinfoot, bbrownin, dtrifiro, weaton
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: ---
Doc Text:
A flaw was found in vLLM. A remote attacker can trigger a Denial of Service (DoS) by submitting a crafted inference request that combines prompt embeddings with sampling penalty parameters. When processed by the penalty handling mechanism, this request causes an unhandled failure that crashes the model-serving engine, disrupting service availability until restarted.
 Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description OSIDB Bzimport 2026-10-06 15:03:57 UTC 
A security flaw has been discovered in vllm-project vLLM up to 0.31.0. This impacts the function get_token_bin_counts_and_mask of the file vllm/model_executor/layers/utils.py of the component Penalty Handler. Performing a manipulation results in denial of service. Remote exploitation of the attack is possible. The exploit has been released to the public and may be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.