Hugging face
D
RISKY
21
Total CVEs
12
Crit / High
7
Medium
0%
Unpatched
0%
Patched
6.7
Avg CVSS
0
CISA KEV
// Data sourced from NVD · CISA KEV · RHSA · GHSA · Debian · OSV · EPSS · Exploit-DB
How we collect data →
CVE Trend by Year
Patch Coverage
✓ 0 patched ✗ 0 unpatched
Intelligence by Audience
PoC Available
0 / 20 shown
CISA KEV (Exploited)
0
Shodan Dorks
ssl:"hugging face" port:443
org:"Hugging face" http.status:200
Top EPSS (Exploitation Probability)
No EPSS data yet for recent CVEs — scores are published by FIRST.org within 24-48h of NVD disclosure.
🎯 Bug Bounty Program: https://huggingface.co/bounty
Quick Facts for Reporting
Total CVEs tracked
21
Critical severity
1
Actively exploited (CISA)
0
Unpatched ratio
0%
Avg CVSS score
6.7
Security grade
D (RISKY)
NVD Data Attribution
Source: NIST National Vulnerability Database (NVD) + GitHub Advisory Database.
Data aggregated by Valters IT Hub · valtersit.com
Last updated: 2026-07-08
Data aggregated by Valters IT Hub · valtersit.com
Last updated: 2026-07-08
Security Risk Score (for investors)
LOW
Vuln Volume
GOOD
Patch Responsiveness
LOW
CISA Exposure
⚠ Not financial advice. Past security incidents do not guarantee future stock performance.
Security News Google News
Recent CVEs
showing 20 of 0
No published CVEs found.
Intel
HQNew York City, United States
Founded2016
CEOClément Delangue
CTOJulien Chaumond
CountryUnited States
LicenseOpen Source
Bug BountyView Program ↗
Trust ScoreD — RISKY
Max CVSS10.0
Last Updated2026-07-08
Top Weaknesses
CWE-502
Deserialization of Untrusted Data
10 occurrences in Hugging face CVEs
(48% of total)
CWE-1333
Unknown Weakness Type
6 occurrences in Hugging face CVEs
(29% of total)
CWE-94
Code Injection
3 occurrences in Hugging face CVEs
(14% of total)
CWE-20
Unknown Weakness Type
1 occurrence in Hugging face CVEs
(5% of total)
CWE-643
Unknown Weakness Type
1 occurrence in Hugging face CVEs
(5% of total)
Company Profile
Hugging Face is a platform and community for machine learning, best known for its open-source Transformers library and model hub that hosts thousands of pre-trained AI models. Its cybersecurity relevance stems from the widespread use of its model repository, which introduces supply chain risks such as malicious models, data poisoning, and vulnerabilities in serialized formats like Pickle. The company also provides tools for model security scanning and has a bug bounty program to address these risks.
Response Metrics
Est. MTTR ⓘ
~500 days
(Critical)
Mean Time To Remediate. Estimates days from CVE disclosure to patch.
Industry avg: ~60d (good) · 100-300d (poor) · 300d+ (critical)
Industry avg: ~60d (good) · 100-300d (poor) · 300d+ (critical)
CVE Trend ⓘ ↗ +500% worse
CVE count 2025 vs 2024:
3 → 18 CVEs
Patch Speed ⓘ Poor (>65% open)
0% of CVEs still unpatched.
Excellent: <20% · Good: 20-40% · Avg: 40-65% · Poor: >65%
Financial Risk Estimate
Est. GDPR Low
€610K
Est. GDPR High
€6.1M
Based on 1 critical CVEs, 0 CISA KEV, 100% unpatched. Model: GDPR Art.83 · not legal advice.
Actions
Embed Badge
Hugging face D valtersit.com