Skip to content
LLMHigh

CVE-2026-115918: vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1…

Occurred
Oct 26, 2026
Published
Oct 26, 2026
Source
NVD CVE

Summary

vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1/responses endpoint allows an authenticated caller to place an attacker-controlled template in chat_template_kwargs[chat_template] when no server template is configured. ChatParams.get_apply_chat_template_kwargs() preserves that value without enforcing trust_request_chat_template, so the request can bypass the intended trust policy and cause CPU or memory exhaustion through Jinja rendering. This issue is fixed in version 0.32.0.

Tags

  • CVE
  • CVE-2026-115918
  • cvss:MEDIUM

Classification

Category confidence: 95%
Severity confidence: 90%

Rationale: CVE describes a security vulnerability in vLLM LLM serving engine allowing template injection leading to resource exhaustion.

Classifier: Qwen/Qwen3.5-35B-A3B-FP8

Visit source