CVE-2025-32381

MEDIUM

mlc-ai xgrammar < 0.1.18 - Denial of Service via Unbounded Grammar Cache

Title source: llm
STIX 2.1

Description

XGrammar is an open-source library for efficient, flexible, and portable structured generation. Prior to 0.1.18, Xgrammar includes a cache for compiled grammars to increase performance with repeated use of the same grammar. This cache is held in memory. Since the cache is unbounded, a system making use of xgrammar can be abused to fill up a host's memory and case a denial of service. For example, sending many small requests to an LLM inference server with unique JSON schemas would eventually cause this denial of service to occur. This vulnerability is fixed in 0.1.18.

References (3)

Core 3
Core References
Issue Tracking, Patch x_refsource_misc
https://github.com/mlc-ai/xgrammar/pull/243
Issue Tracking, Patch x_refsource_misc
https://github.com/vllm-project/vllm/pull/16283

Scores

CVSS v3 6.5
EPSS 0.0035
EPSS Percentile 57.8%
Attack Vector NETWORK
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

CISA SSVC

Vulnrichment
Exploitation none
Automatable no
Technical Impact partial

Details

CWE
CWE-770
Status published
Products (2)
mlc-ai/xgrammar < 0.1.18
pypi/xgrammar 0 - 0.1.18PyPI
Published Apr 09, 2025
Tracked Since Feb 18, 2026