Deserialization of Untrusted Data Affecting rhelai1/instructlab-nvidia-rhel9 package, versions *


Severity

Recommended
0.0
high
0
10

Based on Red Hat Enterprise Linux security rating.

Threat Intelligence

EPSS
0.87% (58th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-RHEL10-RHELAI1INSTRUCTLABNVIDIARHEL9-9898080
  • published29 Apr 2025
  • disclosed27 Jan 2025

Introduced: 27 Jan 2025

CVE-2025-24357  (opens in a new tab)
CWE-502  (opens in a new tab)

How to fix?

There is no fixed version for RHEL:10 rhelai1/instructlab-nvidia-rhel9.

NVD Description

Note: Versions mentioned in the description apply only to the upstream rhelai1/instructlab-nvidia-rhel9 package and not the rhelai1/instructlab-nvidia-rhel9 package as distributed by RHEL. See How to fix? for RHEL:10 relevant fixed versions and status.

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.

CVSS Base Scores

version 3.1