CVE-2025-62426

Published: Nov 21, 2025Modified: Jun 17, 2026

6.5

Vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Exploitability: 2.8 / Impact: 3.6

Source: security-advisories@github.com (Secondary)

Description

vLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before 0.11.1, the /v1/chat/completions and /tokenize endpoints allow a chat_template_kwargs request parameter that is used in the code before it is properly validated against the chat template. With the right chat_template_kwargs parameters, it is possible to block processing of the API server for long periods of time, delaying all other requests. This issue has been patched in version 0.11.1.

Affected (3)

Products: Vllm: Vllm

1 product

Configuration A

3 vulnerable

Vulnerable Software	Affected Versions
Vllm Vllm	From 0.5.5 to 0.11.1
Vllm Vllm	Version 0.11.1 rc0
Vllm Vllm	Version 0.11.1 rc1

Related CWEs

Allocation of Resources Without Limits or Throttling

The product allocates a reusable resource or group of resources on behalf of an actor without imposing any restrictions on the size or number of resources that can be allocated, in violation of the intended security policy for that actor.

References (5)

https://github.com/vllm-project/vllm/blob/2a6dc67eb520ddb9c4138d8b35ed6fe6226997fb/vllm/entrypoints/chat_utils.py#L1602-L1610

Source: security-advisories@github.com

Product

https://github.com/vllm-project/vllm/blob/2a6dc67eb520ddb9c4138d8b35ed6fe6226997fb/vllm/entrypoints/openai/serving_engine.py#L809-L814

Source: security-advisories@github.com

Product

https://github.com/vllm-project/vllm/commit/3ada34f9cb4d1af763fdfa3b481862a93eb6bd2b

Source: security-advisories@github.com

Patch

https://github.com/vllm-project/vllm/pull/27205

Source: security-advisories@github.com

Issue Tracking

https://github.com/vllm-project/vllm/security/advisories/GHSA-69j4-grxj-j64p

Source: security-advisories@github.com

Vendor Advisory

Timeline

No history available yet.