PatchSiren

PatchSiren cyber security CVE debrief

CVE-2026-69112 Hugging Face CVE debrief

CVE-2026-69112 is a path traversal vulnerability in Hugging Face Accelerate versions through 1.14.0. The load_checkpoint_in_model and load_checkpoint_and_dispatch functions do not properly sanitize weight_map entries from sharded checkpoint indexes. This allows attackers to supply relative paths with ../ sequences or absolute paths to read arbitrary files. Additionally, pointing shard entries at named pipes can cause indefinite blocking and denial of service.

Vendor
Hugging Face
Product
Accelerate
CVSS
MEDIUM 6.9
CISA KEV
Not listed in stored evidence
Original CVE published
2026-08-10
Original CVE updated
2026-09-16
Advisory published
2026-08-10
Advisory updated
2026-09-16

Who should care

Defenders responsible for Hugging Face Accelerate deployments should assess exposure and prioritize verification and remediation efforts. This includes reviewing access to sensitive files and named pipes, monitoring for exploitation attempts, and ensuring that Hugging Face Accelerate is updated to a secure version. Additionally, security teams should conduct a thorough risk assessment to identify potential impacts and implement compensating controls for

Why it matters

CVE-2026-69112 is a path traversal vulnerability in Hugging Face Accelerate that allows attackers to read arbitrary files or cause denial of service. Defenders should prioritize verifying and updating Hugging Face Accelerate, reviewing access to sensitive files and named pipes, and monitoring for exploitation attempts.

  • Potential unauthorized file access
  • Potential denial of service through named pipe manipulation
  • Need for verification of Hugging Face Accelerate versions and configurations
  • Priority for updating to a secure version

Technical summary

The load_checkpoint_in_model and load_checkpoint_and_dispatch functions in Hugging Face Accelerate through 1.14.0 are vulnerable to path traversal attacks. Attackers can manipulate weight_map entries from sharded checkpoint indexes to access arbitrary files or cause denial of service by pointing shard entries at named pipes. This vulnerability allows attackers to supply relative paths with ../ sequences or absolute paths to read arbitrary files. Additionally, pointing shard entries at named pipes can cause indefinite blocking and denial of service. Defenders should prioritize verifying and updating Hugging Face Accelerate to a secure version, reviewing and restricting access to sensitive files and named pipes,

Defensive priority

Defenders should prioritize verifying and updating Hugging Face Accelerate to a secure version, reviewing and restricting access to sensitive files and named pipes, and monitoring for potential exploitation attempts.

Recommended defensive actions

  • Verify and update Hugging Face Accelerate to a secure version
  • Review and restrict access to sensitive files and named pipes
  • Monitor for potential exploitation attempts
  • Perform vulnerability scanning to identify exposed systems
  • Review system logs for suspicious activity
  • Implement additional security measures such as network segmentation
  • Conduct a thorough risk assessment to identify potential impacts

Evidence notes

The CVE record and NVD entry provide details on the vulnerability, including its description and CVSS score. However, the corpus does not establish versions, exploitation, impact, or remediation beyond vendor statements.

Sources and references

Verified primary and authoritative sources

  • CVE-2026-69112 CVE Program record

    Publisher, destination, and source semantics verified

    URL: https://www.cve.org/CVERecord?id=CVE-2026-69112

    CVE Program - Official CVE Program record with source-provided CVE metadata.

  • CVE-2026-69112 NVD vulnerability detail

    Publisher, destination, and source semantics verified

    URL: https://nvd.nist.gov/vuln/detail/CVE-2026-69112

    NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.

Supplemental references

Methodology and review provenance

AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.