Vulnerability · Published Sep 16, 2026 · Assigned by VulnCheck
CVE-2026-92786LightGBM through 4.7.0 Out-of-Bounds Write via Crafted Model
High 7.8
In plain English
A high-severity out-of-bounds write in LightGBM from lightgbm-org. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 8% of all known vulnerabilities).
Official description
LightGBM through 4.7.0 fails to validate child and split array values when parsing text models, allowing attackers to write out-of-bounds memory during SHAP prediction. Attackers can craft malicious model files with invalid node references that trigger out-of-bounds writes at attacker-chosen offsets in the leaf_depth_ buffer during feature contribution computation.
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| lightgbm-org | LightGBM | 0 – through 4.7.0 |
Weakness: CWE-787 · Vector: CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
References
- GitHub Issue #7357 issue-tracking
- https://github.com/lightgbm-org/LightGBM product
- RecomputeLeafDepths writes at the bit-inverted child value with no bounds check technical-description
- child and split arrays are parsed with only a length check technical-description
- VulnCheck Advisory: LightGBM through 4.7.0 Out-of-Bounds Write via Crafted Model third-party-advisory