CVE-2025-61620
Description
A flaw was found in the server implementation of vLLM, where the handling of Jinja templates does not properly validate user-supplied input through the chat_template and chat_template_kwargs parameters. When a specially crafted template is processed, it can trigger excessive looping or recursion inside the Jinja engine, consuming large amounts of CPU and memory. This can cause the server to become unresponsive or crash, resulting in a denial-of-service (DoS) condition for applications using vLLM.
Statement
The Red Hat Product Security team has assessed the severity of this vulnerability as Moderate, as it requires authenticated access or the ability to supply templates to the vLLM server. Successful exploitation allows an attacker to exhaust system resources by submitting maliciously crafted Jinja templates that trigger excessive CPU and memory usage. The vulnerability’s root cause is the lack of proper validation and sandboxing of user-supplied template data, which can lead to denial of service (DoS) conditions affecting the availability of services built on vLLM.
Mitigation
Mitigation for this issue is either not available or the currently available options don't meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base or stability.
Common Vulnerability Scoring System (CVSS) Score Details
Info alert:Important note
CVSS scores for open source components depend on vendor-specific factors (e.g. version or build chain). Therefore, Red Hat's score and impact rating can be different from NVD and other vendors. Red Hat remains the authoritative CVE Naming Authority (CNA) source for its products and services (see Red Hat classifications).
CVSS v3 Score Breakdown
| Red Hat | NVD | cve.org | |
|---|---|---|---|
| Base Score | 6.5 | N/A | N/A |
| Attack Vector | Network | N/A | N/A |
| Attack Complexity | Low | N/A | N/A |
| Privileges Required | Low | N/A | N/A |
| User Interaction | None | N/A | N/A |
| Scope | Unchanged | N/A | N/A |
| Confidentiality | None | N/A | N/A |
| Integrity Impact | None | N/A | N/A |
| Availability Impact | High | N/A | N/A |
Vector
Red Hat: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Understanding the Weakness (CWE)
Availability
Technical Impact: DoS: Crash, Exit, or Restart; DoS: Resource Consumption (CPU); DoS: Resource Consumption (Memory); DoS: Resource Consumption (Other)
If an attacker can trigger the allocation of the limited resources, but the number or size of the resources is not controlled, then the most common result is denial of service. This would prevent valid users from accessing the product, and it could potentially have an impact on the surrounding environment, i.e., the product may slow down, crash due to unhandled errors, or lock out legitimate users. For example, a memory exhaustion attack against an application could slow down the application as well as its host operating system.
Access Control,Other
Technical Impact: Bypass Protection Mechanism; Other
In some cases it may be possible to force the product to "fail open" in the event of resource exhaustion. The state of the product -- and possibly the security functionality - may then be compromised.
Acknowledgements
Red Hat would like to thank DarkLight1337, Ga_ryo, Isotr0py, and keymoon for reporting this issue.
Frequently Asked Questions
Not sure what something means? Check out our Security Glossary.
Want to get errata notifications? Sign up here.
