Lucene search
+L

707 matches found

CVE
CVE
added 2025/02/07 7:59 p.m.299 views

CVE-2025-25183

CVE-2025-25183 affects vLLM (prefix cache) where malicious inputs can trigger Python 3.12’s hash(None) behaving as a predictable constant, enabling hash collisions in the prefix cache. This may allow cache entries created from one prompt to be reused for another, causing unintended behavior in re...

2.6CVSS6.8AI score0.00187EPSS
SaveExploits0References3Affected Software1
Vulnrichment
Vulnrichment
added 2025/02/07 7:59 p.m.9 views

CVE-2025-25183 vLLM using built-in hash() from Python 3.12 leads to predictable hash collisions in vLLM prefix cache

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. Prefix caching makes use of Python's built-i...

2.6CVSS3.5AI score0.00187EPSS
SaveExploits0References3
OSV
OSV
added 2025/02/07 7:59 p.m.13 views

CVE-2025-25183 vLLM using built-in hash() from Python 3.12 leads to predictable hash collisions in vLLM prefix cache

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. Prefix caching makes use of Python's built-i...

2.6CVSS4.3AI score
SaveExploits0References5
Positive Technologies
Positive Technologies
added 2025/02/06 12:0 a.m.6 views

PT-2025-6000 · Vllm +1 · Vllm +1

Name of the Vulnerable Software and Affected Versions: vLLM versions prior to 0.7.2 Description: Maliciously constructed statements can lead to hash collisions, resulting in cache reuse, which can interfere with subsequent responses and cause unintended behavior. The issue arises from the use of...

2.6CVSS3.4AI score0.00187EPSS
SaveExploits0References15
Chainguard
Chainguard
added 2025/01/27 8:50 p.m.13 views

GHSA-RH4J-5RHW-HR54 vulnerabilities

Vulnerabilities for packages: tritonserver-backend-vllm, py3.10-vllm-cuda-11.8...

5.3AI score
SaveExploits0
Github Security Blog
Github Security Blog
added 2025/01/27 8:50 p.m.33 views

vllm: Malicious model to RCE by torch.load in hf_model_weights_iterator

Description The vllm/modelexecutor/weightutils.py implements hfmodelweightsiterator to load the model checkpoint, which is downloaded from huggingface. It use torch.load function and weightsonly parameter is default value False. There is a security warning on...

8.8CVSS7.9AI score0.00707EPSS
SaveExploits0References8Affected Software1
Snyk
Snyk
added 2025/01/27 6:42 p.m.10 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 via the hfmodelweightsiterator process due to the usage of the torch.load function with the weightsonly parameter...

8.8CVSS7.8AI score0.00707EPSS
SaveExploits0References2
Chainguard
Chainguard
added 2025/01/27 6:15 p.m.45 views

CVE-2025-24357 vulnerabilities

Vulnerabilities for packages: tritonserver-backend-vllm, py3.10-vllm-cuda-11.8...

8.8CVSS5.1AI score0.00707EPSS
SaveExploits0
NVD
NVD
added 2025/01/27 6:15 p.m.74 views

CVE-2025-24357

vLLM is a library for LLM inference and serving. vllm/modelexecutor/weightutils.py implements hfmodelweightsiterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weightsonly parameter defaults to False. When torch.load loads malicious...

8.8CVSS0.00707EPSS
SaveExploits0References4
OSV
OSV
added 2025/01/27 6:15 p.m.8 views

PYSEC-2025-58

vLLM is a library for LLM inference and serving. vllm/modelexecutor/weightutils.py implements hfmodelweightsiterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weightsonly parameter defaults to False. When torch.load loads malicious...

8.8CVSS7.9AI score0.00707EPSS
SaveExploits0References4
Cvelist
Cvelist
added 2025/01/27 5:38 p.m.72 views

CVE-2025-24357 vLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

vLLM is a library for LLM inference and serving. vllm/modelexecutor/weightutils.py implements hfmodelweightsiterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weightsonly parameter defaults to False. When torch.load loads malicious...

7.5CVSS0.00707EPSS
SaveExploits0References4
CVE
CVE
added 2025/01/27 5:38 p.m.313 views

CVE-2025-24357

The CVE-2025-24357 issue centers on vLLM’s hf_model_weights_iterator (vllm/model_executor/weight_utils.py) which loads checkpoints via torch.load with weights_only defaulting to False. If malicious pickle data is unpickled, arbitrary code could execute on the host. This vulnerability is highlight...

8.8CVSS7.7AI score0.00707EPSS
SaveExploits0References4Affected Software1
OSV
OSV
added 2025/01/27 5:38 p.m.11 views

CVE-2025-24357 vLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

vLLM is a library for LLM inference and serving. vllm/modelexecutor/weightutils.py implements hfmodelweightsiterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weightsonly parameter defaults to False. When torch.load loads malicious...

7.5CVSS7.5AI score
SaveExploits0References6
CNNVD
CNNVD
added 2025/01/27 12:0 a.m.52 views

vLLM 代码问题漏洞

vLLM is a vLLM open source high throughput and memory efficient inference and service engine for LLM. A code issue vulnerability exists in versions of vLLM prior to v0.7.0 that stems from arbitrary code execution during unpickling when torch.load loads malicious pickle data...

8.8CVSS7.3AI score0.00707EPSS
SaveExploits0References4
Positive Technologies
Positive Technologies
added 2025/01/27 12:0 a.m.17 views

PT-2025-5340 · Vllm · Vllm

Name of the Vulnerable Software and Affected Versions: vLLM versions prior to 0.7.0 Description: The issue concerns the vLLM library, specifically the vllm/model executor/weight utils.py file, which implements hf model weights iterator to load model checkpoints downloaded from Hugging Face. It...

8.8CVSS7.3AI score0.00707EPSS
SaveExploits0References14
Chainguard
Chainguard
added 2024/11/23 3:31 a.m.23 views

GHSA-QXRP-VHVM-J765 vulnerabilities

Vulnerabilities for packages: nemo...

5.2AI score
SaveExploits0
Chainguard
Chainguard
added 2024/11/23 3:31 a.m.20 views

GHSA-WRFC-PVP9-MR9G vulnerabilities

Vulnerabilities for packages: nemo...

5.2AI score
SaveExploits0
Chainguard
Chainguard
added 2024/11/23 3:31 a.m.62 views

GHSA-HXXF-235M-72V3 vulnerabilities

Vulnerabilities for packages: nemo...

5.2AI score
SaveExploits0
Chainguard
Chainguard
added 2024/11/22 10:15 p.m.43 views

CVE-2024-11393 vulnerabilities

Vulnerabilities for packages: nemo...

8.8CVSS7.7AI score0.02918EPSS
SaveExploits1
Chainguard
Chainguard
added 2024/11/22 10:15 p.m.106 views

CVE-2024-11394 vulnerabilities

Vulnerabilities for packages: nemo...

8.8CVSS7.7AI score0.02435EPSS
SaveExploits1
Rows per page
Query Builder