CVE-2026-86288: Out-of-Bounds Read in ModelCloud GPTQModel
CVE-2026-86288 is an out-of-bounds read vulnerability in ModelCloud GPTQModel versions 7.0, 7.1, and 7.2.0. It occurs in the Triton dequantization kernel component, specifically in the file gptqmodel/nn_modules/qlinear/tritonv2.py, due to improper handling of the g_idx argument. The vulnerability can be exploited remotely and has been publicly disclosed. Upgrading to version 7.3.0 resolves the issue.
AI Analysis
Technical Summary
This vulnerability in ModelCloud GPTQModel up to version 7.2.0 allows an attacker to cause an out-of-bounds read by manipulating the g_idx argument in the Triton dequantization kernel code (gptqmodel/nn_modules/qlinear/tritonv2.py). The flaw can be triggered remotely. The issue is fixed in version 7.3.0 with patch 877c732f7d7dccd56a729844c6a5bd20f3aa8bb1.
Potential Impact
An attacker can remotely exploit this vulnerability to cause an out-of-bounds read, which may lead to information disclosure or application instability. The CVSS 4.0 base score is 5.3 (medium severity), reflecting network attack vector, low complexity, no privileges required, and user interaction needed with low impact on confidentiality, integrity, and availability.
Mitigation Recommendations
Upgrade ModelCloud GPTQModel to version 7.3.0 or later, which contains the official fix for this vulnerability. No other mitigation is specified or required.
CVE-2026-86288: Out-of-Bounds Read in ModelCloud GPTQModel
Description
CVE-2026-86288 is an out-of-bounds read vulnerability in ModelCloud GPTQModel versions 7.0, 7.1, and 7.2.0. It occurs in the Triton dequantization kernel component, specifically in the file gptqmodel/nn_modules/qlinear/tritonv2.py, due to improper handling of the g_idx argument. The vulnerability can be exploited remotely and has been publicly disclosed. Upgrading to version 7.3.0 resolves the issue.
CVSS v4.0
Score 5.3medium
Affected software
pkg:github/modelcloud/GPTQModelcpe:2.3:a:modelcloud:gptqmodel:*:*:*:*:*:*:*:*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
This vulnerability in ModelCloud GPTQModel up to version 7.2.0 allows an attacker to cause an out-of-bounds read by manipulating the g_idx argument in the Triton dequantization kernel code (gptqmodel/nn_modules/qlinear/tritonv2.py). The flaw can be triggered remotely. The issue is fixed in version 7.3.0 with patch 877c732f7d7dccd56a729844c6a5bd20f3aa8bb1.
Potential Impact
An attacker can remotely exploit this vulnerability to cause an out-of-bounds read, which may lead to information disclosure or application instability. The CVSS 4.0 base score is 5.3 (medium severity), reflecting network attack vector, low complexity, no privileges required, and user interaction needed with low impact on confidentiality, integrity, and availability.
Mitigation Recommendations
Upgrade ModelCloud GPTQModel to version 7.3.0 or later, which contains the official fix for this vulnerability. No other mitigation is specified or required.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulDB
- Date Reserved
- 2026-09-06T15:40:50.945Z
- Cvss Version
- 4.0
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 6a9e77e3acd9273b493fb3c2
Added to database: 09/07/2026, 08:37:55 UTC
Last enriched: 09/07/2026, 08:53:10 UTC
Last updated: 09/07/2026, 08:53:10 UTC
Views: 3
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.