CVE-2025-62426: vLLM Vulnerable to DoS via Chat Completion Request Parameters

# CVE-2025-62426: vLLM Vulnerable to DoS via Chat Completion Request Parameters

https://gcdradar.vercel.app/incidents/cve-2025-62426-vllm-vulnerable-to-dos-via-chat-completion-re · AIR-2025-0376 · closed 2025-11-20

- Lane: Patch now
- Severity: 2 (Disclosed vulnerability)
- Fix status: Fix status unknown
- First reported: 2025-11-20
- AI role: Attack on AI
- Products: vLLM
- CVE ids: CVE-2025-62426
- CVSS: 6.5

## Summary

Improper validation of chat template arguments in vLLM endpoints allows attackers to block API server processing for extended periods. The issue affects chat completion and tokenization requests.

## What to do

No action recorded; read the sources below and the vendor's advisory.

## Latest updates

- 2025-11-21: CVE-2025-62426: vLLM is an inference and serving engine for large language models (LLMs). (https://nvd.nist.gov/vuln/detail/CVE-2025-62426)
- 2025-11-20: CVE-2025-62426: vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs` (https://github.com/advisories/GHSA-69j4-grxj-j64p)

## Sources

- 2025-11-20 GitHub Advisory Database: CVE-2025-62426: vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs` https://github.com/advisories/GHSA-69j4-grxj-j64p
- 2025-11-21 NVD: CVE-2025-62426: vLLM is an inference and serving engine for large language models (LLMs). https://nvd.nist.gov/vuln/detail/CVE-2025-62426

Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.