---
id: CVE-2026-73559
title: vLLM is an inference and serving engine for large language models
summary: >-
  vLLM is an inference and serving engine for large language models. From 0.19.0
  until 0.26.0, the /v1/completions CompletionRequest.prompt field in
  vllm/entrypoints/openai/completion/protocol.py accepts an unbounded list[str]
  or list[list…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-400
vendor: vllm
product: vllm
affected:
  - 'vllm >= 0.19.0, < 0.26.0'
patched:
  - vllm 0.26.0
published: '2026-08-13'
updated: '2026-09-09'
sourceUpdated: '2026-09-09T20:58:37.713'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-73559'
references:
  - url: >-
      https://github.com/vllm-project/vllm/commit/675f4295cdfe0d870471c2b51bfeca3a68a9569e
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/pull/47845'
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/releases/tag/v0.26.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-87x5-vmc3-756j
    label: security-advisories@github.com
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-87x5-vmc3-756j
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-73559'
  - url: 'https://github.com/advisories/GHSA-87x5-vmc3-756j'
  - url: 'https://github.com/vllm-project/vllm'
tags:
  - nvd
  - ghsa
  - pip
  - osv
epss: 0.00548
epssPercentile: 0.43638
aliases:
  - GHSA-87x5-vmc3-756j
  - PYSEC-2026-3704
ecosystem: pip
ingestedAt: '2026-08-13T19:05:04.705Z'
---

## Overview

vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.26.0, the /v1/completions CompletionRequest.prompt field in vllm/entrypoints/openai/completion/protocol.py accepts an unbounded list[str] or list[list[int]], prompt_to_seq() in vllm/renderers/inputs/preprocess.py and OnlineRenderer.preprocess_completion() in vllm/renderers/online_renderer.py expand every element, and vllm/entrypoints/openai/completion/serving.py creates one engine generator and response slot per prompt, allowing an authenticated API client to exhaust CPU, memory, async scheduling capacity, engine request slots, and response buffering with one request. This issue is fixed in version 0.26.0.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-73559)

Affected packages:

- `vllm >= 0.19.0, < 0.26.0`

Patched in:

- `vllm 0.26.0`

Source: https://github.com/advisories/GHSA-87x5-vmc3-756j