---
id: CVE-2026-109228
title: >-
  SGLang through 0.5.21 contains a resource budget bypass vulnerability that
  allows remote attackers to override GLM-V video processing limits by supplying
  fps, max_frames, max_tokens_per_frame and max_image_tokens in the video_url
  object.…
summary: >-
  SGLang through 0.5.21 contains a resource budget bypass vulnerability that
  allows remote attackers to override GLM-V video processing limits by supplying
  fps, max_frames, max_tokens_per_frame and max_image_tokens in the video_url
  object.…
severity: high
cvss: 7.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-770
published: '2026-10-12'
updated: '2026-10-12'
sourceUpdated: '2026-10-12T14:17:12.207'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-109228'
references:
  - url: 'https://github.com/sgl-project/sglang'
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/entrypoints/openai/protocol.py#L592-L594
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/multimodal/processors/glm4v.py#L160-L167
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/multimodal/processors/glm4v.py#L400
    label: disclosure@vulncheck.com
  - url: >-
      https://github.com/sgl-project/sglang/blob/v0.5.21/python/sglang/srt/parser/conversation.py#L691
    label: disclosure@vulncheck.com
  - url: 'https://github.com/sgl-project/sglang/issues/43705'
    label: disclosure@vulncheck.com
  - url: >-
      https://www.vulncheck.com/advisories/sglang-through-0.5.21-resource-budget-bypass-via-glm-v-video-parameters
    label: disclosure@vulncheck.com
tags:
  - nvd
ingestedAt: '2026-10-12T14:50:34.122Z'
---

## Overview

SGLang through 0.5.21 contains a resource budget bypass vulnerability that allows remote attackers to override GLM-V video processing limits by supplying fps, max_frames, max_tokens_per_frame and max_image_tokens in the video_url object. Attackers can set these values arbitrarily high so glm4v.py decodes every frame of high-resolution video at full size, exhausting worker memory and denying service.

## Remediation

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