Resource Exhaustion Affecting vllm-cu130-wheels package, versions <0.29.0-r0


Severity

Recommended
0.0
medium
0
10

Snyk's Security Team recommends NVD's CVSS assessment. Learn more

Threat Intelligence

EPSS
0.33% (25th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-MINIMOSLATEST-VLLMCU130WHEELS-20376757
  • published2 Oct 2026
  • disclosed26 Sept 2026

Introduced: 26 Sep 2026

NewCVE-2026-100651  (opens in a new tab)
CWE-400  (opens in a new tab)

How to fix?

Upgrade Minimos:latest vllm-cu130-wheels to version 0.29.0-r0 or higher.

NVD Description

Note: Versions mentioned in the description apply only to the upstream vllm-cu130-wheels package and not the vllm-cu130-wheels package as distributed by Minimos. See How to fix? for Minimos:latest relevant fixed versions and status.

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' (multimodal) payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied token_ids, and GenerateRequest.token_ids (vllm/entrypoints/serve/disagg/protocol.py) is not checked against model_config.max_model_len. For multimodal processors that report skip_prompt_length_check=True (for example Nemotron Parse, Whisper, and FireRedLID), InputProcessor._validate_prompt_len() returns immediately for both encoder and decoder prompts, so an overlong prompt becomes an EngineCoreRequest and reaches the worker input-batch copy into a fixed max_model_len-wide NumPy row. A client able to reach the endpoint on an affected model configuration can therefore submit an overlong token_ids list to trigger a worker failure and denial of service. Fixed in 0.29.0.

CVSS Base Scores

version 3.1