Lucene search
+L

16 matches found

osv_pysec
osv_pysec
•added 2026/06/29 11:50 a.m.•32 views

PYSEC-2026-567 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Impacted Environments This issue ONLY impacts environments using the PyNcclPipe KV cache transfer integration with the V0 engine. No other configurations are affected. Summary vLLM supports the use of the PyNcclPipe class to establish a peer-to-peer communication domain for data transmission...

9.8CVSS8.2AI score0.00964EPSS
SaveExploits1References8
PyPA
PyPA
•added 2026/06/29 11:50 a.m.•35 views

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Impacted EnvironmentsThis issue ONLY impacts environments using the PyNcclPipe KV cache transfer integration with the V0 engine. No other configurations are affected. SummaryvLLM supports the use of thePyNcclPipeclass to establish a peer-to-peer communication domain for data transmission between...

9.8CVSS8.3AI score0.00964EPSS
SaveExploits1References8Affected Software1
BDU FSTEC
BDU FSTEC
•added 2026/03/20 12:00 a.m.•34 views

PyNcclPipe Class Code Execution

The vulnerability of the PyNcclPipe class in the library for working with Large Language Models LLMs like vLLM is related to deficiencies in the deserialization mechanism. Exploiting this vulnerability allows a remote attacker to execute arbitrary code...

10CVSS8.5AI score0.00964EPSS
SaveExploits1References5Affected Software1
EUVD
EUVD
•added 2025/10/03 8:07 p.m.•44 views

EUVD-2025-15950

Malicious code in bioql PyPI...

9.8CVSS7.9AI score0.00964EPSS
SaveExploits1References6
NVD
NVD
•added 2025/05/20 6:15 p.m.•110 views

CVE-2025-47277

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...

9.8CVSS0.00964EPSS
SaveExploits1References4
Snyk
Snyk
•added 2025/05/20 6:04 p.m.•21 views

Deserialization of Untrusted Data

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Deserialization of Untrusted Data through the PyNcclPipe service if it is in use with the V0 engine. An attacker can execute arbitrary code on the...

9.8CVSS8.2AI score0.00964EPSS
SaveExploits1References2
Github Security Blog
Github Security Blog
•added 2025/05/20 6:04 p.m.•59 views

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Impacted Environments This issue ONLY impacts environments using the PyNcclPipe KV cache transfer integration with the V0 engine. No other configurations are affected. Summary vLLM supports the use of the PyNcclPipe class to establish a peer-to-peer communication domain for data transmission...

9.8CVSS8.3AI score0.00964EPSS
SaveExploits1References9Affected Software1
osv_ghsa
osv_ghsa
•added 2025/05/20 6:04 p.m.•33 views

GHSA-HJQ4-87XH-G4FV vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Impacted Environments This issue ONLY impacts environments using the PyNcclPipe KV cache transfer integration with the V0 engine. No other configurations are affected. Summary vLLM supports the use of the PyNcclPipe class to establish a peer-to-peer communication domain for data transmission...

9.8CVSS8.3AI score0.00964EPSS
SaveExploits1References9
Vulnrichment
Vulnrichment
•added 2025/05/20 5:32 p.m.•17 views

CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

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...

9.8CVSS7.6AI score0.00964EPSS
SaveExploits1References4
Mend
Mend
•added 2025/05/20 5:32 p.m.•19 views

CVE-2025-47277

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...

9.8CVSS7.6AI score0.00964EPSS
SaveExploits1References10
Cvelist
Cvelist
•added 2025/05/20 5:32 p.m.•98 views

CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

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...

9.8CVSS0.00964EPSS
SaveExploits1References4
CVE
CVE
•added 2025/05/20 5:32 p.m.•378 views

CVE-2025-47277

vLLM (versions 0.6.5–0.8.4) is affected only when using the PyNcclPipe KV cache transfer integration with the V0 engine. The issue stems from the PyTorch TCPStore binding defaulting to ALL interfaces; a workaround constrained the store to a private interface, and as of version 0.8.5 vLLM now bind...

9.8CVSS7.6AI score0.00964EPSS
SaveExploits1References4Affected Software1
osv_cve
osv_cve
•added 2025/05/20 5:32 p.m.•39 views

CVE-2025-47277 vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

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...

9.8CVSS7.6AI score0.00964EPSS
SaveExploits1References6
Rapid7 Vulnerability Database (full)
Rapid7 Vulnerability Database (full)
•added 2025/05/20 12:00 a.m.•8 views

CVE-2025-47277: Deserialization of Untrusted Data

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...

9.8CVSS7.7AI score0.00964EPSS
SaveExploits1References1
CNNVD
CNNVD
•added 2025/05/20 12:00 a.m.•101 views

vLLM code issues and vulnerabilities

vLLM is a vLLM open source high throughput and memory efficient inference and service engine for LLM. A code issue vulnerability exists in vLLM versions 0.6.5 through 0.8.4, which stems from PyNcclPipe KV cache transfers not properly limiting the scope of TCPStore interface access...

9.8CVSS7.6AI score0.00964EPSS
SaveExploits1References4
Positive Technologies
Positive Technologies
•added 2025/04/03 12:00 a.m.•3 views

PT-2025-54993

Уязвимость класса PyNcclPipe библиотеки для работы с большими языковыми моделями LLM vLLM связана с недостатками механизма десериализации. Эксплуатация уязвимости может позволить нарушителю, действующему удаленно, выполнить произвольный код...

10CVSS5.8AI score
SaveExploits0References5
Rows per page
Query Builder