CVE-2026-73557Network attack vector

Sparse-validation bypass via process-global flag race

Published Aug 13, 2026 · Updated Aug 13, 2026

Race condition in vLLM 0.20.2rc0 through 0.26.0rc1 allows remote attackers to bypass sparse-tensor validation via concurrent prompt parts. safe_load_prompt_embeds saves, enables, and restores PyTorch's process-global invariant flag without synchronization, letting one executor thread disable another thread's guard before tensor.to_dense. The default-off enable_prompt_embeds setting is required; service interruption and memory corruption remain conditional, and reliable code execution is unproven.

CVSS severity6.3
Medium
EPSS probability0.25%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

Race condition in vLLM 0.20.2rc0 through 0.26.0rc1 allows remote attackers to bypass sparse-tensor validation via concurrent prompt parts. safe_load_prompt_embeds saves, enables, and restores PyTorch's process-global invariant flag without synchronization, letting one executor thread disable another thread's guard before tensor.to_dense. The default-off enable_prompt_embeds setting is required; service interruption and memory corruption remain conditional, and reliable code execution is unproven.

The record

CVE
CVE-2026-73557
Published
Aug 13, 2026
Updated
Aug 13, 2026
Vendor
vLLM Project
Product
vLLM
Classifications
CWE-362, T1499
Attack vector
network
Privileges
unauthenticated

Timeline

How it unfolded

  1. Aug 13, 2026CVE publishedPublication date reported by the CVE source.
  2. Aug 13, 2026Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=>= 0.20.2rc0, < 0.26.0

What conditions does exploitation require?

Attack vectornetwork
Required privilegesunauthenticated

What is affected?

vLLM Project · vLLMversion=>= 0.20.2rc0, < 0.26.0

Published CVSS scores

6.3GitHub, Inc.CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesunauthenticated
Known exploitationUnconfirmed
Public exploitUnconfirmed

Weakness, pattern, technique

CWE-362Concurrent Execution using Shared Resource with Improper Synchronization ('Race Condition')
T1499Endpoint Denial of Service

Public exploit references

No public exploit references are available in this record.

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo