256 published AI models are associated with this package: 23 name it in a requirements or pyproject file, 9 load it or import it directly, and 229 are inferred from the runtime stack their declared library pulls in. Only 18 state a version requirement, so this is not a count of affected models.
cvss
7.1
high
CVSS v3.1 base score out of 10; GitHub rates it MODERATE
epss
0.19%
low
chance of exploitation in 30 days, 8th percentile of all CVEs
xyz score
not scored
CyberXYZ composite out of 10
fig. 01 — GHSA-4j2p-28q2-5m79, the advisory selected below
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.
// cvss v3.1 vector
CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:H
Attack vector
Local
Attack complexity
Low
Privileges required
None
User interaction
Required
Scope
Unchanged
Confidentiality
High
Integrity
None
Availability
High
Checked 2026-11-06 at 19:42 UTC. The most recent advisory here was published 2026-08-10. Updated continuously from NVD, GHSA, OSV and CNA feeds.