---
id: CVE-2026-63764
title: >-
  LMDeploy through 0.14.0, fixed in commit 03c3130, contains a server-side
  request forgery (SSRF) vulnerability in the _load_http_url function within the
  connection.py media handler, where the private-IP guard validates only the
  original U…
summary: >-
  LMDeploy through 0.14.0, fixed in commit 03c3130, contains a server-side
  request forgery (SSRF) vulnerability in the _load_http_url function within the
  connection.py media handler, where the private-IP guard validates only the
  original U…
severity: high
cvss: 8.6
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N'
cwe:
  - CWE-918
vendor: internlm
product: lmdeploy
affected:
  - lmdeploy <= 0.14.0
published: '2026-07-21'
updated: '2026-09-17'
sourceUpdated: '2026-09-17T18:16:57.477'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-63764'
references:
  - url: >-
      https://github.com/InternLM/lmdeploy/commit/03c313006d17cc3feae86b633c44206a997c44db
    label: disclosure@vulncheck.com
  - url: 'https://github.com/InternLM/lmdeploy/issues/4761'
    label: disclosure@vulncheck.com
  - url: 'https://github.com/InternLM/lmdeploy/pull/4734'
    label: disclosure@vulncheck.com
  - url: >-
      https://www.vulncheck.com/advisories/lmdeploy-server-side-request-forgery-via-http-redirect-bypass
    label: disclosure@vulncheck.com
  - url: 'https://github.com/InternLM/lmdeploy/issues/4761'
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
tags:
  - nvd
  - cve.org
  - exploit-available
exploitAvailable: true
ssvc:
  exploitation: poc
  automatable: 'yes'
  technicalImpact: partial
  timestamp: '2026-07-23T14:12:09.324972Z'
epss: 0.00511
epssPercentile: 0.41141
ingestedAt: '2026-09-17T18:25:15.978Z'
---

## Overview

LMDeploy through 0.14.0, fixed in commit 03c3130, contains a server-side request forgery (SSRF) vulnerability in the _load_http_url function within the connection.py media handler, where the private-IP guard validates only the original URL without re-validating hosts after HTTP redirects. An unauthenticated attacker can submit a crafted image_url to the chat completions endpoint pointing to an attacker-controlled host that returns a redirect to a private IP or cloud-metadata endpoint, causing the server to follow the redirect and expose internal service content through the model pipeline.

## Affected

- `lmdeploy <= 0.14.0`

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.