Sign in Sign up
JA · EN · FR
SecPulse
Categories
CVE-2026-22807 high CVSS 8.8

Code Injection in vllm (CVE-2026-22807)

Summary

Code injection in vllm (CVE-2026-22807). Successful exploitation can lead to full system takeover. Exploitable via auto_map. Mitigation: upgrade to 0.14.0 or later.

Original description

The description as published by the source.

Summary

vLLM loads Hugging Face auto_map dynamic modules during model resolution without gating on trust_remote_code, allowing attacker-controlled Python code in a model repo/path to execute at server startup.

Impact

An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load.
This happens before any request handling and does not require API access.

Affected Versions

All versions where vllm/model_executor/models/registry.py resolves auto_map entries with try_get_class_from_dynamic_module without checking trust_remote_code (at least current main).

Details

During model resolution, vLLM unconditionally iterates auto_map entries from the model config and calls try_get_class_from_dynamic_module, which delegates to Transformers’ get_class_from_dynamic_module and executes the module code.

This occurs even when trust_remote_code is false, allowing a malicious model repo to embed code in a referenced module and have it executed during initialization.

Relevant code

  • vllm/model_executor/models/registry.py:856 — auto_map resolution
  • vllm/transformers_utils/dynamic_module.py:13 — delegates to get_class_from_dynamic_module, which executes code

Fixes

Credits

Reported by bugbunny.ai

Source: NIST National Vulnerability Database NVD GitHub Advisory

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-2026-22807 has been found in vllm.
Attackers can target a specific entry point like auto_map over the network to misuse the product.
Successful exploitation can lead to full system takeover. CVSS score: 8.8/10.
What to do: upgrade vllm to 0.14.0 or later.
If unsure, ask your IT team or search "vllm CVE-2026-22807" on the vendor's site.

CVE-2026-22807 (vllm) — CWE-94 / CVSS v3 8.8
Attack vector: remote (network-reachable) / unauthenticated
Attack surface: auto_map / trust_remote_code / try_get_class_from_dynamic_module / main
Patched: 0.14.0 — 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

Arbitrary code execution (CWE-94) exists in vllm. Attackers reach the vulnerable code path via auto_map without authentication.

📍 Affected scope

Check the vendor advisory for the affected versions of vllm. Attack surface: auto_map / trust_remote_code / try_get_class_from_dynamic_module / main.

🔥 Severity

Severity: High (CVSS 8.8/10). Successful exploitation can lead to full system takeover.

🔧 How to fix

Update to 0.14.0.

🛡️ 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 request patterns matching this CVE's known IOCs. 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
    Apply patch patch 
    Upgrade vllm to 0.14.0

    ステージング環境で 0.14.0 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を含めること。

  3. 3
    Post-deployment verification verify 
    Confirm patched version is live in production

    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。本番ではアクセスログ・WAF でアラート再発が無いか継続監視。

Affected packages

pip vllm
[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.14.0"}]}]
PyPI vllm
[{"type":"ECOSYSTEM","events":[{"introduced":"0.10.1"},{"fixed":"0.14.0"}]}]

References

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