Skip to content
Fencer Vulns
high 7.5

CVE-2025-24357 json

vLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator

published 2025-01-27 21 mo ago modified 2026-08-07 3 mo ago also EUVD-2025-0192 , GHSA-rh4j-5rhw-hr54 , PYSEC-2025-58

cvss
7.5
version 3.1, CNA
cisa kev
Not listed
not in CISA's catalog
epss
0.9%
EPSS 0.9% chance of exploitation in 30 days, percentile 59%
public exploits
0
none known

Summary

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.

Patch / fix

  • pypi vllm 0.7.0
  • vllm-project vllm 0.7.0
  • vllm vllm 0.7.0

Vendor advisories

What the vendors themselves say about their products: affected, fixed or not affected, and what to do.

red hat product security vex CVE-2025-24357 updated cvss 7.5

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

  • Red Hat Enterprise Linux AI (RHEL AI) bootc-intel-rhel9, bootc-nvidia-rhel9, bootc-aws-nvidia-rhel9, bootc-gcp-nvidia-rhel9, bootc-ibm-nvidia-rhel9, instructlab-intel-rhel9, +2 more
    affected: workaround, no fix available, no fix planned “Will not fix”
  • Red Hat Enterprise Linux AI (RHEL AI) bootc-azure-amd-rhel9, instructlab-amd-rhel9
    not affected: vulnerable code not present

Red Hat CSAF VEX © Red Hat, Inc., from https://security.access.redhat.com/data/csaf/v2/vex/; changed: reduced to product statuses and remediations. CC BY 4.0 · csaf

CISA SSVC

Exploitation
none
Automatable
no
Technical impact
total
Assessed
21 mo ago

Exploit prediction

0.89% chance of exploitation in the next 30 days, higher than 59% of all scored CVEs (FIRST EPSS, 2026-11-08).

CVSS scores

assigned by version score vector
CNA 3.1 7.5 CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H
GitHub 3.1 7.5 CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H
NVD 3.1 8.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
OSV 3.1 8.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Affected packages

ecosystem package affected fixed in
pypi vllm
<0.7.0
0.7.0

Affected products

  • vllm-project vllm

Classification

Weakness
  • CWE-502 Deserialization of Untrusted Data

References

Sources

Fencer Vulns keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.

CPEs and fixed versions use Anchore's correction of NVD's data: Added CPE configurations because not yet analyzed by NVD.