Skip to main content

CVE-2026-109227: Allocation of Resources Without Limits or Throttling in sgl-project sglang

0
High
Published: 10/12/2026 (10/12/2026, 15:32:18 UTC)
Source: CVE Database V5
Vendor/Project: sgl-project
Product: sglang

Description

SGLang versions up to 0.5.21 have a vulnerability in the preprocess_video function of qwen_vl.py that allows unbounded memory allocation. This occurs because video frames are decoded at full source resolution before applying any pixel limits. Remote attackers can exploit this by submitting high-resolution videos to the /v1/chat/completions endpoint, causing concurrent requests to exhaust worker memory and result in denial of service.

CVSS v4.0

Score 8.7high

Attack Vector
Network
Attack Complexity
Low
Attack Requirements
None
Privileges Required
None
User Interaction
None
Vuln. Confidentiality
None
Vuln. Integrity
None
Vuln. Availability
High
Subsq. Confidentiality
None
Subsq. Integrity
None
Subsq. Availability
None
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N

Affected software

sgl-project

sglang

Affected versions
>=0 <=0.5.21
sglang
pkg:pypi/sglang
Affected versions
>=0 <=0.5.21

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

AILast updated: 10/12/2026, 14:33:29 UTC

Technical Analysis

CVE-2026-109227 is an unbounded memory allocation vulnerability in SGLang through version 0.5.21. The flaw exists in the preprocess_video function within qwen_vl.py, which decodes video frames at full source resolution prior to enforcing max_pixels limits. This allows remote attackers to submit high-resolution videos to the /v1/chat/completions API endpoint on Qwen-VL deployments, leading to excessive memory consumption. When multiple such requests occur concurrently, the worker memory can be exhausted, causing a denial of service condition.

Potential Impact

Successful exploitation results in denial of service by exhausting the memory resources of the worker handling video processing. This can disrupt availability of the affected service, potentially impacting all users relying on the /v1/chat/completions endpoint for video input processing.

Mitigation Recommendations

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, consider limiting the size or resolution of video inputs accepted by the /v1/chat/completions endpoint or implementing request throttling to reduce the risk of memory exhaustion.

Pro Console: star threats, build custom feeds, automate alerts via Slack, email & webhooks.Upgrade to Pro

Technical Details

Data Version
5.2
Assigner Short Name
VulnCheck
Date Reserved
2026-10-12T13:46:32.999Z
Cvss Version
4.0
State
PUBLISHED

Threat ID: 6accec5f2cdf04f6569f8106

Added to database: 10/12/2026, 14:19:11 UTC

Last enriched: 10/12/2026, 14:33:29 UTC

Last updated: 10/12/2026, 21:47:19 UTC

Views: 14

Community Reviews

0 reviews

Crowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.

Sort by
Loading community insights…

Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.

Actions

PRO

Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.

Please log in to the Console to use AI analysis features.

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

Breach by OffSeqOFFSEQFRIENDS — 25% OFF

Check if your credentials are on the dark web

Instant breach scanning across billions of leaked records. Free tier available.

Scan now
OffSeq TrainingCredly Certified

Lead Pen Test Professional

Technical5-day eLearningPECB Accredited
View courses