Servers and applications
GHSA-v5gm-qgmv-gc6c
Out-of-range `stop_token_ids` with `min_tokens` can kill vLLM EngineCore
- Severity
- Medium 6.5
- Published
- 12 Sept 2026
- Last changed
- 12 Sept 2026
- Added to UpHeron
- 3 Nov 2026
- Fixed in
- 0.29.0
- Weakness
- CWE-129
- Attack
- Over the network · Needs a user account · No user action needed
- Attacker can
- Take it down
- CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H- Sources
- Project advisories (GitHub)
Description
vLLM accepts user-controlled stop_token_ids on the OpenAI-compatible /v1/completions and /v1/chat/completions APIs, but only validates that the values are integers. It does not validate that each token id is within the model vocabulary/logits range.
When min_tokens > 0, vLLM uses stop_token_ids as logits indices to suppress stop tokens until the minimum generation length is reached. An out-of-range token id reaches a CUDA indexing operation and triggers a device-side assert, causing EngineCore to enter a fatal state and the API server to fail.
An authenticated API user can send a single malformed completion request that causes denial of service:
- The request returns 500 Internal Server Error.
- EngineCore reports a fatal error.
- Subsequent requests fail until the service is restarted.
The demonstrated impact is service-level DoS through a CUDA device-side assertion.
stop_token_ids is accepted as list[int], and SamplingParams checks only the type:
Affected software
| Software | Affected versions |
|---|---|
vLLMServers and applications · github:vllm-project/vllm |
|
About vLLM
open-source software for large language model inference
- Docker Hub pulls
- 44,195,225Docker Hub, 15 Nov 2026
- GitHub stars
- 96,943GitHub, 15 Nov 2026
- Forks
- 24,836GitHub, 15 Nov 2026
- Latest release
- v0.33.0rc3GitHub, 15 Nov 2026
- Released
- 2 Nov 2026GitHub
- License
- Apache Software License 2.0Wikidata
- First published on GitHub
- 9 Feb 2023GitHub
Website Source code Docker image
Vulnerabilities in vLLM
- In our list
- 169
- Published in the last 12 months
- 150
- Critical, last 12 months
- 6
- High, last 12 months
- 63
- Known exploited
- 0
- CVE-2026-130686vLLM 0.6.0 before 0.33.0 Resource Exhaustion via Multi-Image RequestsHigh 8.7 18 Nov 2026
- CVE-2026-130684vLLM before 0.33.0 Path Traversal via speculators_config verifier.name_or_pathMedium 5.3 18 Nov 2026
- CVE-2026-130685vLLM 0.28.0 before 0.33.0 Unauthenticated DoS via Out-of-Range Prompt Token IDHigh 8.7 18 Nov 2026
- GHSA-xgh8-qwjh-q853/v1/systemone bypasses the --trust-request-chat-template gateMedium 6.5 18 Nov 2026
- CVE-2026-129356vllm-project vllm Common Prefix single_type_kv_cache_manager.py race conditionLow 2.3 15 Nov 2026
- CVE-2026-126394vLLM 0.30.0 before 0.31.0 Resource Exhaustion via OpenPangu Video Backend num_framesMedium 6.9 13 Nov 2026
References
- GitHub Advisory Database https://github.com/advisories/GHSA-v5gm-qgmv-gc6c
- github.com https://github.com/vllm-project/vllm/security/advisories/GHSA-v5gm-qgmv-gc6c