Server crash or code execution via sparse tensor
Published Nov 21, 2025 · Updated Nov 24, 2025
Out-of-bounds write in the vLLM Completions API 0.10.2 through 0.11.0 allows remote attackers to crash the server or execute code. renderer.py decodes a user-supplied prompt embedding, loads it with torch.load while sparse invariant checks are disabled, and calls to_dense on unchecked indices, causing an out-of-bounds memory write. Access to the Completions API is required; a successful write crashes the server process, while reliable code execution has not been demonstrated.
Summary
What happened
Out-of-bounds write in the vLLM Completions API 0.10.2 through 0.11.0 allows remote attackers to crash the server or execute code. renderer.py decodes a user-supplied prompt embedding, loads it with torch.load while sparse invariant checks are disabled, and calls to_dense on unchecked indices, causing an out-of-bounds memory write. Access to the Completions API is required; a successful write crashes the server process, while reliable code execution has not been demonstrated.
The record
- CVE
- CVE-2025-62164
- Published
- Nov 21, 2025
- Updated
- Nov 24, 2025
- Vendor
- vLLM Project
- Product
- vLLM
- Classifications
- CWE-787, CWE-20, CWE-502, CWE-123, T1190
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Nov 21, 2025CVE publishedPublication date reported by the CVE source.
- Nov 24, 2025Record 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.
- Affected versionversion=>= 0.10.2, < 0.11.1
What conditions does exploitation require?
What is affected?
Published CVSS scores
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.
Weakness, pattern, technique
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