All vulnerabilities

Servers and applications

GHSA-xgh8-qwjh-q853

/v1/systemone bypasses the --trust-request-chat-template gate

Severity
Medium 6.5
Published
18 Nov 2026
Last changed
18 Nov 2026
Added to UpHeron
18 Nov 2026
Fixed in
0.34.0
Weakness
Protection mechanism failure (CWE-693)
Attack
Over the network · Needs a user account · No user action needed
Attacker can
Change data
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:N
Sources
Project advisories (GitHub)

Description

In affected vLLM releases, POST /v1/systemone accepted request-level chat_template_kwargs and passed them to the shared HF renderer without calling validate_chat_template(). This let callers provide a chat template even when the server’s default trust setting disallowed request-supplied templates.

The trust gate is OnlineRenderer.validate_chat_template(). It is invoked by the two callers (render_chat, render_responses), not by the shared renderer both of them call (preprocess_chat). /v1/systemone reaches preprocess_chat directly:

StructuredDecisionRequest.chat_template_kwargs protocol.py:42 -> ServingStructuredDecisions.create_decision serving.py:110 -> NextTokenStrategy.read -> _read_request strategies.py:261 -> ReadPromptRequest.chat_template_kwargs protocol.py:95 -> OnlineRenderer.preprocess_chat online_renderer.py:752 -> HfRenderer -> tokenizer.apply_chat_template(...)

From vllm-project/vllm's security advisory. Shortened; the full text is there.

Affected software

SoftwareAffected versions
vLLMServers and applications · github:vllm-project/vllm
  • From 0.32.0rc0 before 0.34.0

About vLLM

open-source software for large language model inference

From Wikidata, CC0 1.0.
Docker Hub pulls
44,195,225Docker Hub, 15 Nov 2026
GitHub stars
96,943GitHub, 15 Nov 2026
Forks
24,836GitHub, 15 Nov 2026
Latest release
v0.33.0rc3GitHub, 15 Nov 2026
Released
2 Nov 2026GitHub
License
Apache Software License 2.0Wikidata
First published on GitHub
9 Feb 2023GitHub

References