NIST deprecates RSA-2048 from 2030. Find every one of them before attackers do.
🐍
🐍 PyPI
Not in CISA KEV
HIGH severity

Accelerate path traversal and denial of service via sharded checkpoint weight_map entriesGHSA-4j2p-28q2-5m79

HIGHFix: huggingface/accelerate#4070

GHSA-4j2p-28q2-5m79 is a high-severity (CVSS 7.1) Path Traversal vulnerability in accelerate. The records list no fixed version yet; check the vendor advisory below and apply the mitigations in the meantime.

Also known asCVE-2026-69112PYSEC-2026-3804
Published
Updated
Affected
1 pkg
Patched
See advisory
Exploits
None indexed
Exploitation data as of Nov 9, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

Proof-of-concept exploit code exists

  • CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.

Exploitation and automatability from CISA’s SSVC triage for GHSA-4j2p-28q2-5m79.

EPSS Exploitation Probability

via FIRST.org ↗
0.2%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs8th percentile — riskier than 8% of all scored CVEsHighest risk
0.00%0.23%0.46%0.69%0.1%0.2%0.2%Sep 26Nov 26Nov 26

Probability of exploitation in the next 30 days, from FIRST.org EPSS.

How urgent is this, really

Where this sits among everything scored

Real-World Exposure

1 pkg affected
🐍accelerate

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Hugging Face Accelerate through 1.14.0 contains a path traversal vulnerability in load_checkpoint_in_model and load_checkpoint_and_dispatch functions that fail to sanitize weight_map entries from sharded checkpoint indexes. Attackers can supply relative paths with ../ sequences or absolute paths to read arbitrary files, or point shard entries at named pipes to cause indefinite blocking and denial of service.

Affected Packages

1 total
EcosystemPackageVulnerable rangeFix
🐍PyPIaccelerateall versionsNo fix

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for accelerate, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

  2. Remediation status

    The records we use list no patched version of accelerate for GHSA-4j2p-28q2-5m79. Check the maintainer's advisory, and where your build allows, override or pin the dependency away from the vulnerable range.

  3. Mitigate without a patch

    Resolve every user-supplied path to its canonical form and reject anything that escapes the intended directory, and run the component under an account that has no read or write access outside the directory it legitimately serves.

Fixing This On Your OS

If you run this on a Linux distribution, patch through your package manager against the distro's own security advisory below — it tracks the exact backported fix for your release, which can ship on a different timeline (and sometimes a different severity) than the upstream project.

Red HatImportant

This is an Important vulnerability in Hugging Face Accelerate that could lead to arbitrary file disclosure and denial of service. While requiring user interaction, an attacker could craft malicious `weight_map` entries in sharded checkpoint indexes to exploit path traversal, impacting the confidentiality and…

Workaround published by Red Hat
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
Source: Red Hat security advisory for GHSA-4j2p-28q2-5m79 (CC BY 4.0)
ProductFixed inAdvisory
Red Hat AI Inference Server 3.4rhaii/vllm-spyre-rhel9:1791858592RHSA-2026:83114
Red Hat AI Inference Server 3.4rhaii/model-opt-cuda-rhel9:1791857110RHSA-2026:83116
Red Hat AI Inference Server 3.4rhaii/vllm-rocm-rhel9:1791856725RHSA-2026:83121
Red Hat AI Inference Server 3.5rhaii/vllm-spyre-rhel9:1791818357RHSA-2026:82371
Red Hat OpenShift AI 2.25rhoai/odh-caikit-tgis-serving-rhel9:1787060100RHSA-2026:65126
Red Hat OpenShift AI 2.25rhoai/odh-caikit-nlp-rhel9:1791448212RHSA-2026:87410
Red Hat OpenShift AI 3.4rhoai/odh-autorag-rhel9:1791557539RHSA-2026:84734
Red Hat OpenShift AI 3.5rhoai/odh-automl-rhel9:1791916705RHSA-2026:86891

Frequently Asked Questions

Hugging Face Accelerate through 1.14.0 contains a path traversal vulnerability in load_checkpoint_in_model and load_checkpoint_and_dispatch functions that fail to sanitize weight_map entries from sharded checkpoint indexes. Attackers can supply relative paths with ../ sequences or absolute paths to read arbitrary files, or point shard entries at named pipes to cause indefinite blocking and denial of service.
O3 Security · Impact-Aware SCA

Is GHSA-4j2p-28q2-5m79 in your dependencies?

Find it across PyPI, including transitive dependencies.

Accelerate path traversal and denial of service via sharded…