Skip to content
Fencer Vulns
high 7.8

CVE-2026-80047 json

Hugging Face Transformers library writes remote code to disk prior to consent check

published 2026-09-01 2 mo ago modified 2026-10-01 38 d ago also GHSA-x9r9-c232-4q39 , PYSEC-2026-4174 , EUVD-2026-69144

cvss
7.8
version 3.1, CNA
cisa kev
Not listed
not in CISA's catalog
epss
0.1%
EPSS 0.1% chance of exploitation in 30 days, percentile 0%
public exploits
0
none known

Summary

A vulnerability in Hugging Face Transformers (versions 4.57.0 to 5.16.1) allows remote Python files to be written to local disk without user consent when using GenerativePreTrainedModel.load_custom_generate(). The function fetches and caches a remote module file before performing the required trust_remote_code consent check, inverting the security model enforced by other code-loading paths (such as AutoConfig, AutoModel, and AutoTokenizer). As a result, attacker‑controlled Python code from custom_generate/generate.py is copied into the user’s ~/.cache/huggingface/modules directory even if the user declines the trust prompt. Although execution is correctly gated, the file write is not reversible and can persist across sessions. This can lead to persistent, unauthorized files on disk and stale cache collisions where cached attacker code may later be executed during trusted model loads. The issue stems from an unconditional file write in dynamic_module_utils.py prior to any trust verification.

Patch / fix

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-2026-80047 updated cvss 6.5

transformers: Hugging Face Transformers: Unauthorized file write via GenerativePreTrainedModel.load_custom_generate()

  • Red Hat Lightspeed Core lightspeed-core/rag-tool-cpu-rhel9, lightspeed-core/lightspeed-stack-rhel9, lightspeed-core/rag-tool-cuda-12.9-rhel9 Red Hat OpenShift Lightspeed openshift-lightspeed/lightspeed-ocp-rag-rhel9, openshift-lightspeed/lightspeed-service-api-rhel9, openshift-lightspeed-tech-preview/lightspeed-rag-tool-rhel9 Red Hat AI Inference Server rhaii/vllm-cpu-rhel9, rhaii/vllm-tpu-rhel9, rhaii/vllm-cuda-rhel9, rhaii/vllm-rocm-rhel9, rhaiis/vllm-cpu-rhel9, rhaiis/vllm-tpu-rhel9, +9 more Red Hat Ansible Automation Platform 2 ansible-automation-platform-25/lightspeed-chatbot-rhel8, ansible-automation-platform-26/lightspeed-chatbot-rhel9, ansible-automation-platform-27/lightspeed-chatbot-rhel9 Red Hat Enterprise Linux AI (RHEL AI) 3 rhelai3/bootc-cuda-rhel9, rhelai3/bootc-rocm-rhel9, rhelai3/bootc-gaudi-rhel9, rhelai3/bootc-aws-cuda-rhel9, rhelai3/bootc-gcp-cuda-rhel9, rhelai3/disk-image-cuda-rhel9, +2 more Red Hat OpenShift AI (RHOAI) rhai/guidellm-rhel9, rhoai/odh-automl-rhel9, rhoai/odh-autorag-rhel9, rhoai/odh-ogx-core-rhel9, rhoai/odh-caikit-nlp-rhel9, rhoai/odh-vllm-gaudi-rhel9, +37 more
    affected: workaround, no fix available “Fix deferred”

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
2 mo ago

Exploit prediction

0.10% chance of exploitation in the next 30 days, higher than 0% of all scored CVEs (FIRST EPSS, 2026-11-08).

CVSS scores

assigned by version score vector
CISA ADP 3.1 7.8 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
CNA 3.1 7.8 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
GitHub 3.1 7.8 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
OSV 3.1 7.8 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Affected packages

ecosystem package affected fixed in
pypi transformers
>=4.49.0, <=5.8.1
 no fix listed

Affected products

  • Hugging Face Transformers

Classification

Weakness
  • CWE-273 Improper Check for Dropped Privileges
  • CWE-367 Time-of-check Time-of-use (TOCTOU) Race Condition
  • CWE-494 Download of Code Without Integrity Check

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.