---
id: CVE-2026-100650
title: >-
  vLLM through 0.29.0 fetches and fully materializes remote or inline media
  before enforcing its documented media controls (the
  VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and
  the per-modality --limit-mm-per-p…
summary: >-
  vLLM through 0.29.0 fetches and fully materializes remote or inline media
  before enforcing its documented media controls (the
  VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and
  the per-modality --limit-mm-per-p…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-400
  - CWE-770
vendor: vllm
product: vllm
affected:
  - vllm <= 0.29.0
published: '2026-09-26'
updated: '2026-10-06'
sourceUpdated: '2026-10-06T19:32:45.337'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-100650'
references:
  - url: >-
      https://github.com/vllm-project/vllm/commit/752a3a504485790a2e8491cacbb35c137339ad34
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv
    label: disclosure@vulncheck.com
  - url: >-
      https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-exhaustion-via-unbounded-media-materialization
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-p6g9-7v3x-m8mv
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100650.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-100650'
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2541909'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-100650'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-100650'
  - url: 'https://access.redhat.com/errata/RHSA-2026:83112'
  - url: 'https://access.redhat.com/errata/RHSA-2026:83120'
  - url: 'https://access.redhat.com/errata/RHSA-2026:83121'
  - url: 'https://access.redhat.com/errata/RHSA-2026:83114'
  - url: 'https://access.redhat.com/errata/RHSA-2026:84734'
tags:
  - nvd
  - cve.org
  - csaf
  - vex
  - red-hat
  - exploit-available
  - osv
  - pip
epss: 0.00808
epssPercentile: 0.56664
exploitAvailable: true
ssvc:
  exploitation: poc
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-09-30T14:49:40.322864Z'
scores:
  nvd: 6.5
  vendor: 7.5
  cna: 6.5
ingestedAt: '2026-09-26T13:33:20.920Z'
aliases:
  - PYSEC-2026-4184
  - GHSA-p6g9-7v3x-m8mv
ecosystem: pip
patched:
  - vllm 0.30.0
---

## Overview

vLLM through 0.29.0 fetches and fully materializes remote or inline media before enforcing its documented media controls (the VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed-audio size cap, default 25 MB, and the per-modality --limit-mm-per-prompt item limits). Across four ingress paths — the shared media-acquisition layer (HTTPConnection.get_bytes()/async_get_bytes()), the chat completions audio_url/base64 path, the batch speech runner, and the Rust frontend POST /tokenize route — the server reads the entire HTTP response body, base64-decodes the inline payload, or spawns one fetch/decode task per media part, and only then applies the limit (or, on some paths, never applies it). A remote attacker can therefore cause the API server or batch-runner process to allocate memory and consume outbound bandwidth proportional to an attacker-chosen body size or media item count before the request is rejected, resulting in pre-inference memory and bandwidth exhaustion (denial of service). The chat and batch surfaces require an API key when one is configured; the Rust frontend /tokenize route is unauthenticated by design. There is no code execution or data disclosure impact.

## Affected

- `vllm <= 0.29.0`

## Remediation

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

## Vendor advisories

- **Red Hat VEX** · Important · affected: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · updated 2026-10-19 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100650.json)
- **RHSA-2026:83112** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-10-14 · [advisory](https://access.redhat.com/errata/RHSA-2026:83112)
- **RHSA-2026:83120** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-10-14 · [advisory](https://access.redhat.com/errata/RHSA-2026:83120)
- **RHSA-2026:83121** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-10-14 · [advisory](https://access.redhat.com/errata/RHSA-2026:83121)
- **RHSA-2026:83114** · Red Hat · fixed in: Red Hat AI Inference Server 3.4 · released 2026-10-14 · [advisory](https://access.redhat.com/errata/RHSA-2026:83114)
- **RHSA-2026:84734** · Red Hat · fixed in: Red Hat OpenShift AI 3.4 · released 2026-10-19 · [advisory](https://access.redhat.com/errata/RHSA-2026:84734)

## Package advisory (CVE-2026-100650)

Affected packages:

- `vllm < 0.30.0`

Patched in:

- `vllm 0.30.0`

Source: https://osv.dev/vulnerability/PYSEC-2026-4184