vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs`
Summary
The /v1/chat/completions and /tokenize endpoints allow a chat_template_kwargs request parameter that is used in the code before it is properly validated against the chat template. With the right chat_template_kwargs parameters, it is possible to block processing of the API server for long periods of time, delaying all other requests
Details
In serving_engine.py, the chat_template_kwargs are unpacked into kwargs passed to chat_utils.py apply_hf_chat_template with no validation on the keys or values in that chat_template_kwargs dict. This means they can be used to override optional parameters in the apply_hf_chat_template method, such as tokenize, changing its default from False to True.
Both serving_chat.py and serving_tokenization.py call into this _preprocess_chat method of serving_engine.py and they both pass in chat_template_kwargs.
So, a chat_template_kwargs like {"tokenize": True} makes tokenization happen as part of applying the chat template, even though that is not expected. Tokenization is a blocking operation, and with sufficiently large input can block the API server's event loop, which blocks handling of all other requests until this tokenization is complete.
This optional tokenize parameter to apply_hf_chat_template does not appear to be used, so one option would be to just hard-code that to always be False instead of allowing it to be optionally overridden by callers. A better option may be to not pass chat_template_kwargs as unpacked kwargs but instead as a dict, and only unpack them after the logic in apply_hf_chat_template that resolves the kwargs against the chat template.
Impact
Any authenticated user can cause a denial of service to a vLLM server with Chat Completion or Tokenize requests.
Fix
Betroffene Software
Klicken Sie auf eine Technologie, um alle zugehörigen Schwachstellen anzuzeigen.
| Plattform | Paket | Anfällige Versionen | Behoben in |
|---|---|---|---|
| Python (PyPI) | vllm | >= 0.5.5, < 0.11.1 | 0.11.1 |
Verweise
- https://github.com/vllm-project/vllm/security/advisories/GHSA-69j4-grxj-j64p
- https://nvd.nist.gov/vuln/detail/CVE-2025-62426
- https://github.com/vllm-project/vllm/pull/27205
- https://github.com/vllm-project/vllm/commit/3ada34f9cb4d1af763fdfa3b481862a93eb6bd2b
- https://github.com/vllm-project/vllm
- https://github.com/vllm-project/vllm/blob/2a6dc67eb520ddb9c4138d8b35ed6fe6226997fb/vllm/entrypoints/chat_utils.py#L1602-L1610
- https://github.com/vllm-project/vllm/blob/2a6dc67eb520ddb9c4138d8b35ed6fe6226997fb/vllm/entrypoints/openai/serving_engine.py#L809-L814
- https://pypi.org/project/vllm
- https://github.com/advisories/GHSA-69j4-grxj-j64p
- https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-2012.yaml
- CVSS-Vektor
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H- Schwäche (CWE)
CWE-770- Plattform
- Python (PyPI)
- Paket
vllm- Veröffentlicht
- 20.11.2025
- Aktualisiert
- 11.10.2026 15:17
Quellen und Lizenzen
OSV / GitHub Advisory Database - PYSEC-2026-2012 - PyPA Advisory Database; CC BY 4.0
OSV / GitHub Advisory Database - GHSA-69j4-grxj-j64p - CC BY 4.0
Schwachstellenhistorie: vllm
Alle veröffentlichten Schwachstellen dieses Pakets (Python (PyPI)), neueste zuerst. Zeigt, ob es ein einmaliger Ausrutscher ist oder ein Paket, das man im Auge behalten muss.
- Einträge insgesamt
- 94
- Erste → letzte
- 09.2024 10.2026
- Ohne Fix
- 5
- In Angriffen ausgenutzt
- 0
Verteilung nach Schweregrad
- Kritisch 9
- Hoch 23
- Mittel 47
- Niedrig 4
- Nicht bewertet 11
Veröffentlichungen pro Jahr
- 2024: 2
- 2025: 30
- 2026: 62
-
vLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs` dieser Eintrag
Erkennen Sie Schwachstellen in Ihrer Anwendung
Mit der DockRay-Integration und aktivierter Softwareinventarisierung prüft DockRay gemeldete Komponentenversionen auf bekannte Schwachstellen und benachrichtigt Sie über bestätigte Treffer. Die Abdeckung hängt von der unterstützten Technologie und den verfügbaren Meldungen ab.