Vulnerability in vllm (CVE-2026-5497, CVSS 7.5)

Summary

Vulnerability in vllm (CVE-2026-5497). Risk of unauthorized operations or information disclosure. Mitigation: upgrade to 0.19.0 or later.

Original description

The description as published by the source.

vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the VideoMediaIO.load_base64() method. When processing video/jpeg data URLs, the method splits the base64 data string on commas to extract individual JPEG frames without enforcing a frame count limit. An attacker can exploit this by crafting a single API request containing thousands of comma-separated base64-encoded JPEG frames in a data URL, causing the server to decode all frames into memory and crash due to excessive memory consumption. This vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.

Source: NIST National Vulnerability Database NVD GitHub Advisory

AI summary

Written by an external AI model and may contain mistakes. Check the original description and references below before acting on it.

vLLMのバージョン0.8.0以降で、VideoMediaIO.load_base64()メソッドによるフレーム数制限のない処理のために、メモリ不足のDoS攻撃を受ける可能性があります。攻撃者は大量のBase64エンコードされたJPEGフレームを含むリクエストを送信してサーバをクラッシュさせることができます。

❓ What is the problem

vLLMの脆弱性はフレーム数制限のないメモリ不足DoS攻撃です。

📍 Affected scope

VideoMediaIO.load_base64()メソッド

🔥 Severity

認証不要で攻撃可能なため、高リスクです。

🔧 How to fix

フレーム数制限を追加してメモリの消費を制限します。

🛡️ Workaround

一時的な回避策は情報なし。

🔍 Detection

大量のJEPGデコードが発生する場合に影響を確認可能です。

Affected packages

pip vllm

[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.19.0"}]}]

PyPI vllm

[{"type":"ECOSYSTEM","events":[{"introduced":"0.8.0"},{"fixed":"0.19.0"}]}]

Summary

Original description

AI summary

Affected packages

References