CVE Enrichment

CVSS
9.8 CRITICAL
EPSS
0.96%
EPSS Percentile
61.08%
CISA KEV
Not listed

Description

vLLM, an inference and serving engine for large language models (LLMs), has an issue in versions 0.6.5 through 0.8.4 that ONLY impacts environments using the `PyNcclPipe` KV cache transfer integration with the V0 engine. No other configurations are affected. vLLM supports the use of the `PyNcclPipe` class to establish a peer-to-peer communication domain for data transmission between distributed nodes. The GPU-side KV-Cache transmission is implemented through the `PyNcclCommunicator` class, while CPU-side control message passing is handled via the `send_obj` and `recv_obj` methods on the CPU side.​ The intention was that this interface should only be exposed to a private network using the IP address specified by the `--kv-ip` CLI parameter. The vLLM documentation covers how this must be limited to a secured network. The default and intentional behavior from PyTorch is that the `TCPStore` interface listens on ALL interfaces, regardless of what IP address is provided. The IP address given was only used as a client-side address to use. vLLM was fixed to use a workaround to force the `TCPStore` instance to bind its socket to a specified private interface. As of version 0.8.5, vLLM limits the `TCPStore` socket to the private interface as configured.

Technical Details

CVSS Version
3.1
CNA / NVD Source
security-advisories@github.com
Published
21-May-2025 01:15
Last Modified
17-Jun-2026 16:27
EPSS Date
2026-10-20
CWE
CWE-502

CVSS Vector

CVSS:3.1 AV:N / AC:L / PR:N / UI:N / S:U / C:H / I:H / A:H

Affected Products / Versions

References

Last enrichment sync: 21-Oct-2026 04:13
Sources: NVD · FIRST EPSS · CISA KEV
Latest successful provider update; providers refresh independently.

Reported Context (1)

CVE (7)

Malware (1)

Threat Actors (2)

MITRE ATT&CK (5)

Vendors (10)

Products (5)

Tools (1)

Countries (4)

Note: Related entities, including threat actors, malware, CVEs, MITRE ATT&CK techniques, vendors, products, tools, countries, and industries, are shown when they appear in the same reporting. Their presence does not necessarily mean they were targeted, compromised, vulnerable, responsible for the activity, or directly involved in the incident.