---
id: CVE-2026-73557
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
  0.20.2rc0 until 0.26.0, safe_load_prompt_embeds in
  vllm/renderers/embed_utils.py uses
  torch.sparse.check_sparse_tensor_invariants, whose process-global save,
  enable,…
severity: medium
cwe:
  - CWE-362
vendor: vllm
product: vllm
affected:
  - 'vllm >= 0.21.0, < 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-2026-73557'
references:
  - url: >-
      https://github.com/vllm-project/vllm/commit/793cf79c89d4049124e756915468ac30318f2e50
    label: security-advisories@github.com
  - url: 'https://github.com/vllm-project/vllm/pull/48583'
    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-pr7f-p5mw-fc87
    label: security-advisories@github.com
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-73557'
  - url: 'https://github.com/advisories/GHSA-pr7f-p5mw-fc87'
  - url: 'https://github.com/vllm-project/vllm'
  - url: 'https://pypi.org/project/vllm'
tags:
  - nvd
  - ghsa
  - pip
  - osv
epss: 0.00404
epssPercentile: 0.31964
aliases:
  - GHSA-pr7f-p5mw-fc87
  - PYSEC-2026-3938
ecosystem: pip
ingestedAt: '2026-09-04T22:28:54.863Z'
---

## Overview

vLLM is an inference and serving engine for large language models. From 0.20.2rc0 until 0.26.0, safe_load_prompt_embeds in vllm/renderers/embed_utils.py uses torch.sparse.check_sparse_tensor_invariants, whose process-global save, enable, and restore state can be raced by concurrent prompt_embeds parts submitted to POST /v1/chat/completions through AsyncMultiModalItemTracker.resolve_items, asyncio.gather, and the default executor, allowing an invalid sparse tensor to reach tensor.to_dense despite the CVE-2025-62164 guard when enable_prompt_embeds is enabled. 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-2026-73557)

Affected packages:

- `vllm >= 0.21.0, < 0.26.0`

Patched in:

- `vllm 0.26.0`

Source: https://github.com/advisories/GHSA-pr7f-p5mw-fc87