Sign in Sign up
JA · EN · FR
SecPulse
Categories
CVE-2026-34755 medium CVSS 6.5

Vulnerability in vllm (CVE-2026-34755)

Summary

Vulnerability in vllm (CVE-2026-34755). Risk of unauthorized operations or information disclosure. Exploitable via POST /v1/chat/completions. Mitigation: upgrade to 0.19.0 or later.

Original description

The description as published by the source.

vLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/media/video.py splits video/jpeg data URLs by comma to extract individual JPEG frames, but does not enforce a frame count limit. The num_frames parameter (default: 32), which is enforced by the load_bytes() code path, is completely bypassed in the video/jpeg base64 path. An attacker can send a single API request containing thousands of comma-separated base64-encoded JPEG frames, causing the server to decode all frames into memory and crash with OOM. This vulnerability is fixed in 0.19.0.

Source: OSV.dev (Open Source Vulnerabilities) NVD GitHub Advisory OSV

Auto-generated summary

Assembled by SecPulse from the source data (CVSS, CWE, affected versions) using templates — no AI involved. Check the original description and references below for exact details.

A vulnerability tracked as CVE-2026-34755 has been found in vllm.
Attackers can target a specific entry point like POST /v1/chat/completions over the network to misuse the product.
Risk of unauthorized operations or information disclosure. CVSS score: 6.5/10.
What to do: upgrade vllm to 0.19.0 or later.
If unsure, ask your IT team or search "vllm CVE-2026-34755" on the vendor's site.

CVE-2026-34755 (vllm) — CWE-770 / CVSS v3 6.5
Attack vector: remote (network-reachable) / no user interaction
Attack surface: POST /v1/chat/completions / num_frames / self.num_frames / data
Patched: 0.19.0 — apply immediately
Plan: 1) Audit SBOM/dependencies, 2) Stage→prod upgrade, 3) Add WAF/proxy monitoring on affected endpoints, 4) Hunt IOCs in logs.
Refs: see the GHSA / vendor advisory / patched release linked on this page.

❓ What is the problem

A vulnerability (CWE-770) exists in vllm. Attackers reach the vulnerable code path via POST /v1/chat/completions.

📍 Affected scope

Check the vendor advisory for the affected versions of vllm. Attack surface: POST /v1/chat/completions / num_frames / self.num_frames / data.

🔥 Severity

Severity: Medium (CVSS 6.5/10). Risk of unauthorized operations or information disclosure.

🔧 How to fix

Update to 0.19.0.

🛡️ Workaround

Until the patch is applied: disable the affected feature, apply WAF rules, or restrict access via network ACLs.

🔍 Detection

Search webserver/proxy logs for unusual POST /v1/chat/completions requests with malformed payloads or SQL meta-characters. Run grep -r 'vllm' . against your dependency files (package-lock.json, requirements.txt, go.sum) to find affected services.

Response actions

Concrete steps and command examples for SOC/SRE teams to execute in order

  1. 1
    Identify exposure identify 
    grep -r 'vllm' . | grep -v node_modules

    リポジトリと本番環境の依存ファイル (package-lock.json / requirements.txt / go.sum / Gemfile.lock 等) で vllm を grep し、稼働しているサービス・バージョンを把握する。

  2. 2
    Hunt for indicators of compromise detect 
    grep 'POST /v1/chat/completions' /var/log/nginx/access.log | grep -E '(unusual_payload|sqli_pattern)'

    アクセスログで POST /v1/chat/completions への異常なリクエスト (不正な認証ヘッダ・SQLメタ文字)を過去 30〜90日分捜索。WAF/SIEM があれば該当パスのアラート発火履歴を確認。

  3. 3
    Apply patch patch 
    Upgrade vllm to 0.19.0

    ステージング環境で 0.19.0 に上げて回帰テスト → 本番反映。回帰テストはアプリの主要ハッピーパスと、Step 2 で見つけた異常検知の続報チェックを含めること。

  4. 4
    Post-deployment verification verify 
    Replay attack against POST /v1/chat/completions on staging to confirm patch closes the vector

    パッチ適用後、ステージングで PoC または同等の悪用パターンを再現して脆弱性が閉じたことを確認。本番では Step 2 と同じログクエリでアラート再発が無いか継続監視。

Affected packages

PyPI vllm
[{"type":"ECOSYSTEM","events":[{"introduced":"0.7.0"},{"fixed":"0.19.0"}]}]
pip vllm
[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.19.0"}]}]

References

We use cookies for analytics and ad measurement. This setting applies to all Game co., Ltd. sites. Details