Is CVE-2026-33626 actively exploited?

Yes. CVE-2026-33626 is listed in the CISA Known Exploited Vulnerabilities (KEV) catalog, which means exploitation in the wild has been confirmed.

Can CVE-2026-33626 be exploited remotely?

Yes. CVE-2026-33626 is exploitable remotely over the network (CVSS attack vector: Network).

What is the CVSS score of CVE-2026-33626?

CVE-2026-33626 has a CVSS score of 7.5 (HIGH).

HOME VULNERABILITIESCVE-2026-33626

HIGHCISA KEVIN THE WILD

CVE-2026-33626

CWE-918Published: April 20, 2026· Updated: Apr 23, 2026

7.5

CVSS v3.1

EPSS:0.03%probability of exploitation in 30 daysPercentile:9.0th

Official Description

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.

NVD Source

Technical Analysis

CVE-2026-33626 can be exploited remotely over the network without requiring physical or adjacent access, significantly expanding the attack surface for threat actors.

The vulnerability requires no privileges and no user interaction, making it a prime target for automated exploitation campaigns and worm-like propagation.

A successful exploit results in complete confidentiality breach (data exposure), with a CVSS base score of 7.5.

CISA has added CVE-2026-33626 to the Known Exploited Vulnerabilities (KEV) catalog, confirming active exploitation in the wild. U.S. federal agencies are required to patch this within the mandated timeframe, and all organizations should treat remediation as urgent.

CVSS v3.1 Vector Breakdown

Exploitability

Attack VectorNetwork

Attack ComplexityLow

Privileges Req.None

User InteractionNone

ScopeUnchanged

Impact

ConfidentialityHigh

IntegrityNone

AvailabilityNone

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N

Affected Vendors & Products

internlm1 product

lmdeploy

Source: NVD CPE · 1 total CPE entries

Exploit & PoC Resources

ACTIVE EXPLOITATIONConfirmed exploitation in the wild

Exploit-DB

Offensive Security exploit database

GitHub PoC Search

Community PoC repositories

Rapid7 / Metasploit

Metasploit module search

PacketStorm

Security advisories & exploits

External links open in a new tab. Always verify in a controlled environment before use.

Official Patches & Advisories

PATCHhttps://github.com/InternLM/lmdeploy/commit/71d64a339edb901e9005358e0633fbbab367d626

PATCHhttps://github.com/InternLM/lmdeploy/pull/4447

ADVISORYhttps://github.com/InternLM/lmdeploy/security/advisories/GHSA-6w67-hwm5-92mq

News & Research Mentioning CVE-2026-33626

LMDeploy CVE-2026-33626 Flaw Exploited Within 13 Hours of Disclosure

The Hacker News· Apr 24, 2026

A high-severity security flaw in LMDeploy, an open-source toolkit for compressing, deploying, and serving large language models (LLMs), has come under active exploitation in the wild less than 13 hours after its public disclosure. The vulnerability, tracked as CVE-2026-33626 (CVSS score: 7.5), relates to a Server-Side Request Forgery (SSRF) vulnerability that could be exploited to access [xlite_meta score:37 src:The Hacker News xlite_fp:47c580a28e062c761560de72eb4fbb72e458458533ae225e644774bdd8bb142d]