UTC

transformers

24 published advisories; version 5.10.0 or later fixes all but 1 of them.

Registry
PyPI
Advisories
24
Latest fix
5.10.0
Exploited in the wild
None listed

Check your projects

shieldpi agent-config checks agent and MCP packages from npm and PyPI, not transformers. Any scanner that reads the OSV database finds it in your lockfiles, for example Google's open-source OSV-Scanner.

osv-scanner scan source -r .

Or check a lockfile against ShieldMap in your browser.

Malicious lookalikes

Packages reported malicious by OpenSSF whose names imitate transformers. Check that you install transformers itself.

All AI lookalikes →
Provenance

Source: github.com/huggingface/transformers, named in the package metadata (not verified)

10 of the last 10 releases of transformers were published with build provenance: a signed record of the repository and commit each was built from, which the registry checks before it accepts the release.

  • 5.23.0
  • 5.22.2
  • 5.22.1
  • 5.22.0
  • 5.21.1
  • 5.21.0
  • 5.20.2
  • 5.20.1
  • 5.20.0
  • 5.19.0

Green: published with provenance. Grey: without.

OpenSSF Scorecard: 6.5 of 10 on 2026-11-02.

No dangerous CI workflows
10/10
CI tokens read-only by default
0/10
Build dependencies pinned
3/10
Branch protection
Not checked
Changes reviewed before merge
9/10
Signed releases
Not checked
Actively maintained
10/10
Security policy
10/10
Full Scorecard →

Provenance from deps.dev (CC BY 4.0); Scorecard from OpenSSF (CDLA-Permissive-2.0). As of 2026-11-18.

Advisories
AdvisorySeverityAffectedFixed inPublished
Hugging Face Transformers downloads custom generation code before trust consentCVE-2026-80047High>= 4.49.0, <= 5.8.1No fix yet2026-09-01
Transformers save_pretrained path traversal allows arbitrary file writes through chat template namesCVE-2026-9856High< 5.10.05.10.02026-08-02
huggingface/transformers: Arbitrary Code Execution During Model Initialization in the LightGlue Model Loading PathCVE-2026-5241High< 5.5.05.5.02026-06-03
HuggingFace transformers vulnerable to remote code executionCVE-2026-4372High< 5.3.05.3.02026-05-26
HuggingFace Transformers allows for arbitrary code execution in the `Trainer` classCVE-2026-1839Medium< 5.0.0rc35.0.0rc32026-04-07
Hugging Face Transformers vulnerable to Regular Expression Denial of Service (ReDoS) in the AdamWeightDecay optimizerCVE-2025-6921Medium< 4.53.04.53.02025-09-23
Hugging Face Transformers library has Regular Expression Denial of ServiceCVE-2025-6051Medium< 4.53.04.53.02025-09-14
Hugging Face Transformers is vulnerable to ReDoS through its MarianTokenizerCVE-2025-6638Medium< 4.53.04.53.02025-09-12
Hugging Face Transformers Regular Expression Denial of Service (ReDoS) vulnerabilityCVE-2025-5197Medium< 4.53.04.53.02025-08-06
Transformers is vulnerable to ReDoS attack through its DonutProcessor classCVE-2025-3933Medium<= 4.51.34.52.12025-07-11
Transformers vulnerable to ReDoS attack through its SETTING_RE variableCVE-2025-3262Medium>= 4.49.0, < 4.51.04.51.02025-07-07
Transformers vulnerable to ReDoS attack through its get_imports() functionCVE-2025-3264Medium< 4.51.04.51.02025-07-07
Transformers's Improper Input Validation vulnerability can be exploited through username injectionCVE-2025-3777Low< 4.52.14.52.12025-07-07
Transformers's ReDoS vulnerability in get_configuration_file can lead to catastrophic backtrackingCVE-2025-3263Medium< 4.51.04.51.02025-07-07
Hugging Face Transformers Regular Expression Denial of ServiceCVE-2025-2099Medium< 4.50.04.50.02025-05-19
Transformers Regular Expression Denial of Service (ReDoS) vulnerabilityCVE-2025-1194Medium< 4.50.04.50.02025-04-29
Transformers Regular Expression Denial of Service (ReDoS) vulnerabilityCVE-2024-12720Medium< 4.48.04.48.02025-03-20
Deserialization of Untrusted Data in Hugging Face TransformersCVE-2024-11394High>= 0, < 4.48.04.48.02024-11-23
Deserialization of Untrusted Data in Hugging Face TransformersCVE-2024-11392High>= 0, < 4.48.04.48.02024-11-23
Deserialization of Untrusted Data in Hugging Face TransformersCVE-2024-11393High>= 0, < 4.48.04.48.02024-11-23
Transformers Deserialization of Untrusted Data vulnerabilityCVE-2024-3568Low< 4.38.04.38.02024-04-10
transformers has a Deserialization of Untrusted Data vulnerabilityCVE-2023-7018High< 4.36.04.36.02023-12-20
transformers has a Deserialization of Untrusted Data vulnerabilityCVE-2023-6730Critical< 4.36.04.36.02023-12-19
transformers has Insecure Temporary FileCVE-2023-2800Medium< 4.30.04.30.02023-05-18

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].