CVE-2025-24357
High
Description
vLLM is a library for LLM inference and serving. vllm/model_executor/weight_utils.py implements hf_model_weights_iterator to load the model checkpoint, which is downloaded from huggingface. It uses the torch.load function and the weights_only parameter defaults to False. When torch.load loads malicious pickle data, it will execute arbitrary code during unpickling. This vulnerability is fixed in v0.7.0.
SeverityHigh
CVSS7.5
Exploit probability—
ProductVllm
PublishedJan 27, 2025
First seenJan 27, 2025
Weakness typeCWE-502Deserialization of untrusted data
How to fix
Update
- vllm vllm0.7.0 or later
- vllm0.7.0 or later
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| vllm vllm | v < 0.7.0 | 0.7.0 |
| vllmPyPI | v < 0.7.0 | 0.7.0 |
Check your version
Choose the product and enter the version you run to see whether it falls in the affected range.
In the Intelligence 360 plan
Attack Surface Management
See the assets tied to your domains, the technologies they run and known vulnerabilities in one list.