Lucene search
+L

641 matches found

Snyk
Snyk
added 2026/01/27 10:49 p.m.2 views

Server-side Request Forgery (SSRF)

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Server-side Request Forgery SSRF via the MediaConnector class. An attacker can access internal network resources and cause system instability or...

7.1CVSS5.9AI score0.00528EPSS
SaveExploits1References2
CVE
CVE
added 2026/01/27 10:1 p.m.27 views

CVE-2026-24779

CVE-2026-24779 is an SSRF vulnerability in vLLM’s MediaConnector. Before version 0.14.1, load_from_url and load_from_url_async fetch media from user-supplied URLs and validate via Python urllib urlparse, while the request is issued with requests/urllib3, whose parsing follows a different standard...

7.1CVSS5.9AI score0.00528EPSS
SaveExploits1References15Affected Software1
ATTACKERKB
ATTACKERKB
added 2026/01/27 10:1 p.m.4 views

CVE-2026-24779

vLLM is an inference and serving engine for large language models LLMs. Prior to version 0.14.1, a Server-Side Request Forgery SSRF vulnerability exists in the MediaConnector class within the vLLM project's multimodal feature set. The loadfromurl and loadfromurlasync methods obtain and process...

7.1CVSS5.9AI score0.00528EPSS
SaveExploits1References4Affected Software1
Vulnrichment
Vulnrichment
added 2026/01/27 10:1 p.m.6 views

CVE-2026-24779 vLLM vulnerable to Server-Side Request Forgery (SSRF) in `MediaConnector`

vLLM is an inference and serving engine for large language models LLMs. Prior to version 0.14.1, a Server-Side Request Forgery SSRF vulnerability exists in the MediaConnector class within the vLLM project's multimodal feature set. The loadfromurl and loadfromurlasync methods obtain and process...

7.1CVSS5.9AI score0.00528EPSS
SaveExploits1References3
Positive Technologies
Positive Technologies
added 2026/01/27 12:0 a.m.6 views

PT-2026-5029

Name of the Vulnerable Software and Affected Versions vLLM versions prior to 0.14.1 Description A Server-Side Request Forgery SSRF exists in the MediaConnector class within the multimodal feature set. The load from url and load from url async functions process media from user-provided URLs using...

7.1CVSS6.8AI score0.00528EPSS
SaveExploits1References29
Chainguard
Chainguard
added 2026/01/23 1:20 a.m.9 views

CVE-2026-22807 vulnerabilities

Vulnerabilities for packages: py3-vllm-cuda-12.4, tritonserver-backend-vllm-cuda-12.9, tritonserver-backend-vllm-cuda-13.0...

9.8CVSS6.6AI score0.00737EPSS
SaveExploits1
Chainguard
Chainguard
added 2026/01/23 1:20 a.m.7 views

GHSA-2PC9-4J83-QJMR vulnerabilities

Vulnerabilities for packages: py3-vllm-cuda-12.4, tritonserver-backend-vllm-cuda-12.9, tritonserver-backend-vllm-cuda-13.0...

5.3AI score
SaveExploits0
NVD
NVD
added 2026/01/21 10:15 p.m.9 views

CVE-2026-22807

vLLM is an inference and serving engine for large language models LLMs. Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path ...

9.8CVSS0.00737EPSS
SaveExploits1References17
Cvelist
Cvelist
added 2026/01/21 9:13 p.m.23 views

CVE-2026-22807 vLLM affected by RCE via auto_map dynamic module loading during model initialization

vLLM is an inference and serving engine for large language models LLMs. Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path ...

8.8CVSS0.00737EPSS
SaveExploits1References4
Snyk
Snyk
added 2026/01/21 4:12 p.m.4 views

Arbitrary Code Injection

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Arbitrary Code Injection via the automap process during model initialization, even when trustremotecode is false. An attacker can execute arbitrary...

9.8CVSS6.3AI score0.00737EPSS
SaveExploits1References2
OSV
OSV
added 2026/01/21 4:12 p.m.4 views

GHSA-2PC9-4J83-QJMR vLLM affected by RCE via auto_map dynamic module loading during model initialization

Summary vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path to execute at server startup. --- Impact An attacker who can influence the model repo/path local directory or remote...

8.8CVSS6AI score0.00737EPSS
SaveExploits1References22
Github Security Blog
Github Security Blog
added 2026/01/21 4:12 p.m.13 views

vLLM affected by RCE via auto_map dynamic module loading during model initialization

Summary vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path to execute at server startup. --- Impact An attacker who can influence the model repo/path local directory or remote...

9.8CVSS5.9AI score0.00737EPSS
SaveExploits1References22Affected Software1
CNNVD
CNNVD
added 2026/01/21 12:0 a.m.14 views

vLLM code injection vulnerability

vLLM is an open-source LLM-based inference and service engine that features high throughput and efficient memory usage. Versions of vLLM from 0.10.1 to 0.14.0 contained a code injection vulnerability. This vulnerability stemmed from the loading of the Hugging Face automap dynamic module during...

9.8CVSS7.4AI score0.00737EPSS
SaveExploits1References4
Positive Technologies
Positive Technologies
added 2026/01/21 12:0 a.m.12 views

PT-2026-3865

Name of the Vulnerable Software and Affected Versions vLLM versions 0.10.1 through 0.13.x Description vLLM is an inference and serving engine for large language models LLMs. The software loads Hugging Face auto map dynamic modules during model resolution without verifying trust remote code. This...

9.8CVSS7.5AI score0.00737EPSS
SaveExploits1References36
Github Security Blog
Github Security Blog
added 2026/01/13 6:44 p.m.11 views

vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions

Summary Users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimension mismatch that results in an unhandled runtime error, leading to complete server termination. Details T...

7.5CVSS6.8AI score0.00403EPSS
SaveExploits1References6Affected Software1
Snyk
Snyk
added 2026/01/10 7:40 a.m.4 views

Allocation of Resources Without Limits or Throttling

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the processimageinput in the idefics3 model implementation. An attacker can cause the server...

7.5CVSS6.8AI score0.00403EPSS
SaveExploits1References2
OSV
OSV
added 2026/01/10 6:39 a.m.7 views

CVE-2026-22773 vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions

vLLM is an inference and serving engine for large language models LLMs. In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimensi...

6.5CVSS6.7AI score0.00403EPSS
SaveExploits1References3
CNNVD
CNNVD
added 2026/01/10 12:0 a.m.6 views

vLLM 安全漏洞

vLLM is vLLM open source a high throughput and memory efficient inference and service engine for LLM. A security vulnerability exists in vLLM version 0.6.4 up to and including version 0.12.0, which stems from the fact that sending specially crafted 1x1 pixel images results in a tensor dimension...

7.5CVSS6.4AI score0.00403EPSS
SaveExploits1References1
OSV
OSV
added 2026/01/08 9:47 p.m.8 views

GHSA-MCMC-2M55-J8JJ vLLM introduced enhanced protection for CVE-2025-62164

Summary The fix here for CVE-2025-62164 is not sufficient. The fix only disables prompt embeds by default rather than addressing the root cause, so the DoS vulnerability remains when the feature is enabled. Details vLLM's pending change attempts to fix the root cause, which is the missing sparse...

8.8CVSS6.7AI score0.00352EPSS
SaveExploits0References3
Chainguard
Chainguard
added 2026/01/07 1:30 a.m.4 views

GHSA-HF3C-WXG2-49Q9 vulnerabilities

Vulnerabilities for packages: tritonserver-backend-vllm...

7AI score
SaveExploits0
Rows per page
Query Builder