CVE-2026-33626: CWE-918: Server-Side Request Forgery (SSRF) in InternLM lmdeploy
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.
AI Analysis
Technical Summary
CVE-2026-33626 is a high-severity SSRF vulnerability in InternLM's lmdeploy toolkit, specifically in versions before 0.12.3. The vulnerability exists in the load_image() function within lmdeploy/vl/utils.py, which fetches images from URLs without validating if the URLs point to internal or private IP addresses. This lack of validation enables attackers to make the server perform unauthorized requests to internal network resources, including cloud metadata services, potentially exposing sensitive information. The vulnerability is addressed by updating to lmdeploy version 0.12.3.
Potential Impact
Successful exploitation allows an unauthenticated attacker to cause the vulnerable server to make arbitrary HTTP requests to internal or private network addresses. This can lead to unauthorized access to sensitive internal resources such as cloud metadata services, which may expose confidential data or credentials. The vulnerability does not impact integrity or availability directly but has a high confidentiality impact.
Mitigation Recommendations
Upgrade lmdeploy to version 0.12.3 or later, where the SSRF vulnerability in the load_image() function is patched. Since no official remediation level or patch link is provided beyond this version update, users should verify the upgrade from the vendor's official release notes or repository. No other mitigations are specified.
CVE-2026-33626: CWE-918: Server-Side Request Forgery (SSRF) in InternLM lmdeploy
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.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
CVE-2026-33626 is a high-severity SSRF vulnerability in InternLM's lmdeploy toolkit, specifically in versions before 0.12.3. The vulnerability exists in the load_image() function within lmdeploy/vl/utils.py, which fetches images from URLs without validating if the URLs point to internal or private IP addresses. This lack of validation enables attackers to make the server perform unauthorized requests to internal network resources, including cloud metadata services, potentially exposing sensitive information. The vulnerability is addressed by updating to lmdeploy version 0.12.3.
Potential Impact
Successful exploitation allows an unauthenticated attacker to cause the vulnerable server to make arbitrary HTTP requests to internal or private network addresses. This can lead to unauthorized access to sensitive internal resources such as cloud metadata services, which may expose confidential data or credentials. The vulnerability does not impact integrity or availability directly but has a high confidentiality impact.
Mitigation Recommendations
Upgrade lmdeploy to version 0.12.3 or later, where the SSRF vulnerability in the load_image() function is patched. Since no official remediation level or patch link is provided beyond this version update, users should verify the upgrade from the vendor's official release notes or repository. No other mitigations are specified.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- GitHub_M
- Date Reserved
- 2026-03-23T14:24:11.617Z
- Cvss Version
- 3.1
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 69e6941f19fe3cd2cd32c48f
Added to database: 4/20/2026, 9:01:19 PM
Last enriched: 4/20/2026, 9:16:04 PM
Last updated: 4/21/2026, 7:07:58 AM
Views: 11
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.