Answer in brief
CVE-2026-80047 records a High severity (CVSS 7.8) vulnerability in Hugging Face Transformers library writes remote code to disk prior to consent check. The current sources do not mark it as known exploited. The current feed maps Hugging Face/Transformers (generic), Hugging Face/Transformers (generic), transformers (pip), transformers (pypi). Check affected ranges and fixed versions before updating.
Analysis pending evidence review
HOL Guard separates source facts from reviewed analysis. See the methodology.
CVSS is 7.8. The current sources do not mark it as known exploited. Treat this as a source-backed prioritization signal, not a statement about your environment.
Analysis status
Analysis pending evidence review
Factual feed record only; HOL analysis is not approved for indexing. Read the methodology.
The current feed maps Hugging Face/Transformers (generic), Hugging Face/Transformers (generic), transformers (pip), transformers (pypi). Check affected ranges and fixed versions before updating.
| Package | Affected range | Fixed version |
|---|---|---|
| Hugging Face/Transformersgeneric | 4.49.0 | Not reported |
| Hugging Face/Transformersgeneric | >=4.57.0 <=5.16.1 | Not reported |
| transformerspip | >=4.49.0,<=5.8.1 | Not reported |
| transformerspypi | >=4.49.0 | Not reported |
Published upstream
Sep 1, 2026
Evidence: source:cvelist:source_dates:source-dates:recordSource modified
Oct 1, 2026
Evidence: source:cvelist:source_dates:source-dates:recordFirst seen by HOL
Sep 1, 2026
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.
Quoted source text, attributed separately from HOL analysis.