HighCVE-2026-80047

    Hugging Face Transformers library writes remote code to disk prior to consent check (CVE-2026-80047)

    Published Last modified Sources: OSV, CVE.org

    Share

    What is affected, what to do

    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).

    The available sources name no fixed version. Check the vendor guidance in the references below.

    Affected
    • Hugging Face Transformers lte 4.57.0..5.16.1

    Not sure whether and how you are affected? Ask a question about this vulnerability

    Metrics

    7.8

    Priority

    XIEM Effective CVSS: highest available CVSS rating from NVD and vendor data.

    Source: Vendor (CNA), CVSS v3

    7.8

    CVSS v3

    Severity High on the scale from 0 to 10.

    Source: OSV, CVE.org

    0.1 %

    EPSS score

    Low — model estimates < 1% exploitation likelihood.

    Source: FIRST EPSS, percentile 0.8

    Show all metrics
    Severity
    High
    Priority (0–325)

    The priority score is a consolidated metric that condenses several individual signals into a single, action-guiding ranking: technical severity (CVSS), exploitation likelihood (EPSS), known-exploited status (KEV) and the context of the affected asset. The aim is to answer the only truly important question in vulnerability management: what first?

    Read in glossary
    63.23 Methodology
    Exploit maturity

    An exploit is code or a technique that specifically uses a concrete vulnerability to force unintended behaviour — such as executing code or escalating privileges. It is the tool that turns a theoretical flaw into a practical attack.

    Read in glossary
    no public PoC known
    XIEM Effective CVSS

    XIEM® is the tiered managed-security operating model from NEOSEC and a registered trademark of NEOSEC GmbH. It extends classic SIEM — the central collection, normalisation and correlation of security-relevant events — with endpoint forensics, network visibility and deception technology, and consolidates them into three building tiers: Sentry, Orchestrate and Command.

    Read in glossary
    7.8 Methodology
    CVSS v3

    CVSS is the globally established, open standard for assessing the technical severity of a vulnerability on a scale from 0 to 10. It describes how easily a flaw can be exploited and what damage it can cause — but not how likely its actual exploitation is. CVSS and the complementary EPSS deliberately answer different questions.

    Read in glossary
    7.8
    EPSS score

    The EPSS score is the absolute model output for an individual vulnerability: the estimated probability of real-world exploitation within the next 30 days, expressed as a value between 0 and 1 (i.e. 0 % to 100 %). A score of 0.42 means „the model estimates a 42 % chance of real-world exploitation within 30 days“.

    Read in glossary
    0.1 %
    Low — model estimates < 1% exploitation likelihood.
    EPSS percentile

    The EPSS percentile is the relative position of a vulnerability within all EPSS-scored CVEs, expressed as a value between 0 and 1. A percentile of 0.90 means „this CVE has a higher EPSS score than 90 % of all other scored CVEs“. Unlike the EPSS score, it is not a probability — but a positioning within a ranking.

    Read in glossary
    0.8
    Low — this CVE sits in the lower 10% of all CVEs scored today.
    KEV status

    The KEV catalogue is a list maintained by the US authority CISA of vulnerabilities for which active exploitation in the wild has been confirmed. It answers not the question „how severe?“ but „is it already being attacked?“ — thereby sharply prioritising patching.

    Read in glossary
    —
    Published
    CISA assessment (SSVC)
    Exploitation: none known · not automatable · Technical impact: total
    Source: CISA Vulnrichment, as of 3 September 2026
    CWE

    CWE is a catalogue of typical weakness types in software and hardware — such as flawed input validation, unsafe memory handling or poor authentication. Where a CVE names a concrete vulnerability in a concrete product, a CWE describes the underlying class of defect.

    Read in glossary
    CWE-273, CWE-367, CWE-494

    Detection rules

    No NEOSEC detection rule is available for this vulnerability yet.

    Classification under NIS2 and CRA

    Rule-based classification from exploitation status and severity. It does not replace a case-by-case assessment.

    NIS2

    NIS2 is EU Directive (EU) 2022/2555 to raise the level of cybersecurity across the Union. It substantially broadens obligations compared with the first NIS directive: more sectors, mandatory risk management, staged reporting duties and personal accountability of senior management.

    Read in glossary
    High technical severity: assess and remediate the vulnerability in vulnerability handling under Art. 21(2)(e) NIS2 and document the decision.
    CRA

    The CRA is EU Regulation (EU) 2024/2847 setting horizontal cybersecurity requirements for „products with digital elements“ — from industrial controllers and software to connected consumer devices. For the first time manufacturers must ensure security across the entire product lifecycle, handle vulnerabilities and provide security updates.

    Read in glossary
    From 11.12.2027, Annex I Part II CRA requires manufacturers of affected products with digital elements to handle vulnerabilities without delay and provide security updates.
    ISO/IEC 27001

    ISO/IEC 27001 is the internationally leading standard for an information security management system (ISMS). It describes a risk-based, process-oriented framework to systematically plan, implement, review and improve information security.

    Read in glossary
    Control A.8.8, management of technical vulnerabilities: assess relevance to your systems, plan remediation and keep evidence.

    Description

    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.

    Source: OSV (Open Source Vulnerabilities), cvelistv5, NVD (NIST)

    Affected products

    The following versions are affected. Older releases of the same product line are also vulnerable unless stated otherwise.

    Affected products
    Affected versionsFixed in
    Hugging FaceTransformers
    • 4.57.0 – ≤ 5.16.1
    • —
    pypitransformers
    • 4.53.2
    • 5.1.0
    • 5.2.0
    • 5.0.0rc2
    • 4.51.2
    • 5.7.0
    • 4.57.6
    • 4.50.0
    • 4.54.0
    • 4.49.0
    • 4.52.2
    • 4.52.0
    • 4.57.5
    • 4.57.2
    • 5.6.2
    • 5.5.3
    • 4.54.1
    • 5.5.4
    • 5.4.0
    • 4.53.1
    • 4.50.3
    • 4.57.4
    • 5.6.1
    • 5.3.0
    • 4.55.3
    • 4.52.1
    • 4.55.2
    • 4.55.0
    • 5.8.1
    • 4.57.1
    • 5.0.0rc3
    • 5.5.0
    • 4.56.0
    • 4.51.0
    • 4.56.2
    • 4.53.3
    • 4.52.3
    • 4.51.3
    • 4.50.2
    • 4.55.4
    • 5.8.0
    • 5.0.0rc0
    • 4.57.0
    • 4.57.3
    • 4.52.4
    • 4.56.1
    • 4.55.1
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —
    • —

    Vendors and products

    Weakness classes (CWE)

    • CWE-273Improper Check for Dropped Privileges

      Base

      The product attempts to drop privileges but does not check or incorrectly checks to see if the drop succeeded.

      Entry at MITRE CWE-273

    • CWE-367Time-of-check Time-of-use (TOCTOU) Race Condition

      Base

      The product checks the state of a resource before using that resource, but the resource's state can change between the check and the use in a way that invalidates the results of the check.

      Entry at MITRE CWE-367

    • CWE-494Download of Code Without Integrity Check

      Base

      The product downloads source code or an executable from a remote location and executes the code without sufficiently verifying the origin and integrity of the code.

      Entry at MITRE CWE-494

    References & sources

    Changes to this CVE

    Who changed this CVE and when: NVD (analysis, affected products, CVSS), the CVE publisher (CNA) and additional providers such as CISA (ADP).

    1. Record modifiedcert.org

      Description added or changed

      Show 2 original changes
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Descriptionchanged
      A vulnerability in Hugging Face Transformers (versions >= 4.49.0 and <= 5.8.1) 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.
      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.
      Affected versions (publisher)added
      —
      Affected value modified. <a href="https://github.com/CVEProject/cvelistV5/blob/main/cves/2026/80xxx/CVE-2026-80047.json">CVE-2026-80047</a>

      Source identifier: cret@cert.org

    2. Record modifiedcert.org

      Description added or changed

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Descriptionchanged
      A vulnerability in Hugging Face Transformers (versions 4.49.0, <= 5.8.1) 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.
      A vulnerability in Hugging Face Transformers (versions >= 4.49.0 and <= 5.8.1) 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.

      Source identifier: cret@cert.org

    3. Record modifiedCVE Program

      Show the original change
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Referenceadded
      —
      https://www.kb.cert.org/vuls/id/456290

      Source identifier: af854a3a-2127-422b-91ae-364da2661108

    4. Receivedcert.org

      Description added or changed

      Show 4 original changes
      Individual changes of this event as published by NVD
      TypeActionBeforeAfter
      Descriptionadded
      —
      A vulnerability in Hugging Face Transformers (versions 4.49.0, <= 5.8.1) 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.
      Affected versions (publisher)added
      —
      New affected value received. <a href="https://github.com/CVEProject/cvelistV5/blob/main/cves/2026/80xxx/CVE-2026-80047.json">CVE-2026-80047</a>
      Referenceadded
      —
      https://github.com/huggingface/transformers
      Referenceadded
      —
      https://kb.cert.org/vuls/id/456290

      Source identifier: cret@cert.org

    Embed badge

    Show the severity and KEV status of this vulnerability on your site – with a link back here. The image is updated hourly.

    Badge preview for CVE-2026-80047
    <a href="https://intel.neosec.eu/en/cve/CVE-2026-80047"><img src="https://intel.neosec.eu/badge/cve/CVE-2026-80047.svg?lang=en" alt="CVE-2026-80047 – NEOSEC Intel" height="20"></a>