# CVE-2026-33626: LMDeploy server-side request forgery in vision-language image loading

Source: https://junglewise.ai/threats/cve-2026-33626-lmdeploy-has-server-side-request-forgery-ssrf-via-vision-language  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-04-20. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-33626 · Severity: high · CVSS 7.5 · Published 2026-04-20

LMDeploy is an inference engine for large language models that supports vision capabilities. The vulnerability allows attackers to send specially crafted requests to cause the server to fetch arbitrary URLs, including cloud credential services and internal systems. This can lead to theft of cloud credentials, access to internal services, and lateral movement within corporate networks.

Technologies: [Internlm Lmdeploy](https://junglewise.ai/threats/technologies/internlm-lmdeploy), [lmdeploy (PyPI)](https://junglewise.ai/threats/technologies/lmdeploy). Vendors: [Internlm](https://junglewise.ai/threats/vendors/internlm), [PyPI](https://junglewise.ai/threats/vendors/pypi).

## Quick answers

**Is CVE-2026-33626 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-33626 being exploited in the wild.

**How severe is CVE-2026-33626?** High severity, with a CVSS score of 7.5, published 20 April 2026.

**Which products and versions are affected?** lmdeploy (PyPI) <= 0.12.2, ECOSYSTEM:introduced=0,last_affected=0.12.2.

**Is there a fix for CVE-2026-33626?** A patch was published on 2026: Fix available in LMDeploy version 0.12.3 and later Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-33626 attacked?** It is attacked over the network.

**Am I affected by CVE-2026-33626?** Check whether you run Internlm Lmdeploy or lmdeploy (PyPI) at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

LMDeploy's vision-language module contains a server-side request forgery (SSRF) vulnerability in the `load_image()` and `encode_image_base64()` functions within `lmdeploy/vl/utils.py`. These functions fetch image URLs via HTTP requests without validating or filtering private/internal IP address ranges. An unauthenticated attacker can send a POST request to `/v1/chat/completions` with a malicious image URL pointing to cloud metadata services (e.g., 169.254.169.254), internal networks, or localhost services. The server will fetch and process these URLs, exposing sensitive data. The root cause is the absence of URL validation and IP blocklist functionality. Patches are available in version 0.12.3 and later.

## Affected products

- LMDeploy LMDeploy before 0.12.3

## Identifiers

- CVE-2026-33626
- GHSA-6w67-hwm5-92mq
- PYSEC-2026-2607

## Timeline

- 2026-04-21: disclosed: Public disclosure via GitHub Security Advisory GHSA-6w67-hwm5-92mq
- 2026-02-04: other: Vulnerability tested and confirmed on main branch
- 2026: patched: Fix available in LMDeploy version 0.12.3 and later

## Related threats

- [LMDeploy URL validation bypass via backslash interpretation](https://junglewise.ai/threats/cve-2026-123818-lmdeploy-is-a-toolkit-for-compressing-deploying-and-serving)
- [InternLM LMDeploy unauthenticated remote code execution in /update_weights](https://junglewise.ai/threats/cve-2026-119706-internlm-lmdeploy-0-9-0-through-0-18-0-contains-an)
- [LMDeploy arbitrary code execution via eval() of untrusted quant_dtype](https://junglewise.ai/threats/cve-2026-33625-lmdeploy-arbitrary-code-execution-via-eval-of-untrusted-quant)
- [LMDeploy remote code execution via pickle deserialization in DistServe](https://junglewise.ai/threats/cve-2025-66455-lmdeploy-remote-code-execution-via-pickle-deserialization-in)
- [LMDeploy SSRF bypass in URL validation](https://junglewise.ai/threats/lmdeploy-ssrf-bypass-in-url-validation-c4a40a8d)