Skip to content
Fencer Intel
low 3.3

CVE-2026-63632 json

ONNX: Heap-Buffer-Overflow READ in Gemm Version Converter Adapter via Undersized Input Shape

published 4mo ago modified last mo. also EUVD-2026-60944 , GHSA-p893-rvq9-2xf9 , PYSEC-2026-3587

cvss
3.3
version 3.1, CNA
cisa kev
Not listed
not in CISA's catalog
epss
0.2%
EPSS 0.2% chance of exploitation in 30 days, percentile 6%
public exploits
0
none known

Summary

Open Neural Network Exchange (ONNX) is an open standard for machine learning interoperability. From 1.3.0 until 1.22.0, onnx.version_converter.convert_version() can perform an out-of-bounds read in Gemm_7_6::adapt_gemm_7_6() in onnx/version_converter/adapters/gemm_7_6.h when a Gemm node has input tensors with fewer than two dimensions because B_shape[1], A_shape[0], or A_shape[1] is accessed without a rank check, potentially causing a process crash during an opset 7 to 6 downgrade. This issue is fixed in version 1.22.0.

Patch / fix

  • pypi onnx 1.22.0
  • linuxfoundation onnx 1.22.0
  • onnx onnx 1.22.0
  • pytorch Azure Linux 3.0 2.2.2-17

azure linux advisories AZL-96728

Azure Linux vulnerability data, Copyright (c) Microsoft Corporation, from https://github.com/microsoft/AzureLinuxVulnerabilityData via OSV, under the MIT licence; changed: normalised to fixed package versions per release. MIT

Is CVE-2026-63632 in your dependencies? Fencer shows which of your repositories are affected as soon as an advisory lands, and opens the fix as a pull request.

Check with Fencer

Vendor advisories

What the vendors themselves say about their products: affected, fixed or not affected, and what to do.

red hat product security vex CVE-2026-63632 updated cvss 3.3

onnx: ONNX: Denial of Service via out-of-bounds read in version converter

  • Red Hat OpenShift AI (RHOAI) rhoai/odh-mlserver-rhel9, rhoai/odh-openvino-model-server-rhel9, rhoai/odh-pipeline-runtime-pytorch-cuda-py312-rhel9, rhoai/odh-pipeline-runtime-pytorch-rocm-py312-rhel9, rhoai/odh-workbench-jupyter-pytorch-cuda-py312-rhel9, rhoai/odh-workbench-jupyter-pytorch-rocm-py312-rhel9, +10 more
    affected: workaround, no fix available “Fix deferred”
  • Red Hat OpenShift AI (RHOAI) rhoai/odh-mlserver-cuda-rhel9
    not affected: vulnerable code not present

Red Hat CSAF VEX © Red Hat, Inc., from https://security.access.redhat.com/data/csaf/v2/vex/; changed: reduced to product statuses and remediations. CC BY 4.0 · csaf

CISA SSVC

Exploitation
Proof of concept is public
Automatable
no
Technical impact
partial
Assessed
2mo ago

Exploit prediction

0.17% chance of exploitation in the next 30 days, higher than 6% of all scored CVEs (FIRST EPSS, 2026-11-11).

Exploits and checks

  • github.com https://github.com/onnx/onnx/security/advisories/GHSA-p893-rvq9-2xf9 referenced PoC, tagged "exploit" by a source, unverified

CVSS scores

assigned by version score vector
CNA 3.1 3.3 CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:L
GitHub 3.1 3.3 CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:L
OSV 3.1 3.3 CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:L

Affected packages

ecosystem package affected fixed in
pypi onnx
>=1.3.0, <1.22.0
1.22.0

Affected products

  • onnx onnx

Classification

Weakness

References

Sources

Fencer Intel keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.

The maintainers of onnx/onnx published their own advisory on GitHub. Reviewed sources take precedence over it.

CPEs and fixed versions use Anchore's correction of NVD's data: Added CPE configurations because not yet analyzed by NVD.