| Summary | LMDeploy is a toolkit for compressing, deploying, and serving large language models. Versions prior to 0.12.3 have a Server-Side Request Forgery (SSRF) vulnerability in LMDeploy's vision-language module. The `load_image()` function in `lmdeploy/vl/utils.py` fetches arbitrary URLs without validating internal/private IP addresses, allowing attackers to access cloud metadata services, internal networks, and sensitive resources. Version 0.12.3 patches the issue. |
|---|---|
| Publication Date | April 21, 2026, 6:16 a.m. |
| Registration Date | April 25, 2026, 4:02 a.m. |
| Last Update | April 23, 2026, 10:39 p.m. |
| CVSS3.1 : HIGH | |
| スコア | 7.5 |
|---|---|
| Vector | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N |
| 攻撃元区分(AV) | ネットワーク |
| 攻撃条件の複雑さ(AC) | 低 |
| 攻撃に必要な特権レベル(PR) | 不要 |
| 利用者の関与(UI) | 不要 |
| 影響の想定範囲(S) | 変更なし |
| 機密性への影響(C) | 高 |
| 完全性への影響(I) | なし |
| 可用性への影響(A) | なし |
| Configuration1 | or higher | or less | more than | less than | |
| cpe:2.3:a:internlm:lmdeploy:*:*:*:*:*:*:*:* | 0.12.3 | ||||