CVE-2026-24747
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
Attack profile
How easy it is to exploit
- Attack vector
- Network
- Complexity
- Low
- Privileges needed
- None
- User interaction
- Required
What an attacker can affect
- Confidentiality
- High
- Integrity
- High
- Availability
- High
Exploit likelihood
What each source reports
CVE List Score used
8.8
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
CVE-2026-24747- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Weakness
- CWE-502, CWE-94, CWE-502
- Updated
- 2026-11-02
EUVDEUVD-2026-4738
8.8
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
EUVD-2026-4738- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Updated
- 2026-11-02
GitHub AdvisoriesGHSA-63cw-57p8-fm3p
8.8
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
Summary
A vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution.
Vulnerability Details
The weights_only=True unpickler failed to properly validate pickle opcodes and storage metadata, allowing:
- Heap memory corruption via
SETITEM/SETITEMSopcodes applied to non-dictionary types - Storage size mismatch between declared element count and actual data in the archive
Impact
An attacker who can convince a user to load a malicious checkpoint file may achieve arbitrary code execution in the context of the victim's process.
Credit
Ji'an Zhou
- Record
GHSA-63cw-57p8-fm3p- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Weakness
- CWE-94, CWE-502
- Updated
- 2026-11-04
NVD
8.8
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
CVE-2026-24747- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Weakness
- CWE-94, CWE-502
- Updated
- 2026-11-02
OSV
8.8
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
CVE-2026-24747- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Weakness
- CWE-502, CWE-94
- Updated
- 2026-11-02
OSVDEBIAN-CVE-2026-24747
8.8
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
DEBIAN-CVE-2026-24747- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Updated
- 2026-11-02
Show 7 moreShow less
OSVGHSA-63cw-57p8-fm3p
8.8
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
Summary
A vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution.
Vulnerability Details
The weights_only=True unpickler failed to properly validate pickle opcodes and storage metadata, allowing:
- Heap memory corruption via
SETITEM/SETITEMSopcodes applied to non-dictionary types - Storage size mismatch between declared element count and actual data in the archive
Impact
An attacker who can convince a user to load a malicious checkpoint file may achieve arbitrary code execution in the context of the victim's process.
Credit
Ji'an Zhou
- Record
GHSA-63cw-57p8-fm3p- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Weakness
- CWE-502, CWE-94
- Updated
- 2026-11-04
OSVPYSEC-2026-1856
8.8
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
Summary
A vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution.
Vulnerability Details
The weights_only=True unpickler failed to properly validate pickle opcodes and storage metadata, allowing:
- Heap memory corruption via
SETITEM/SETITEMSopcodes applied to non-dictionary types - Storage size mismatch between declared element count and actual data in the archive
Impact
An attacker who can convince a user to load a malicious checkpoint file may achieve arbitrary code execution in the context of the victim's process.
Credit
Ji'an Zhou
- Record
PYSEC-2026-1856- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Updated
- 2026-11-02
OSVPYSEC-2026-2286
8.8
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
PYSEC-2026-2286- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Updated
- 2026-11-02
OSVUBUNTU-CVE-2026-24747
8.8
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
UBUNTU-CVE-2026-24747- CVSS 3.1
- 8.8 High
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H - Updated
- 2026-11-02
OSVAZL-75293 No score
CVE-2026-24747 affecting package pytorch for versions less than 2.2.2-11
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
AZL-75293- Updated
- 2026-11-05
OSVAZL-75588 No score
CVE-2026-24747 affecting package pytorch for versions less than 2.0.0-14
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
AZL-75588- Updated
- 2026-11-02
OSVBIT-pytorch-2026-24747 No score
PyTorch Vulnerable to Remote Code Execution via Untrusted Checkpoint Files
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's weights_only unpickler allows an attacker to craft a malicious checkpoint file (.pth) that, when loaded with torch.load(..., weights_only=True), can corrupt memory and potentially lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- Record
BIT-pytorch-2026-24747- Updated
- 2026-11-02
Affected products
Affected packages
| Package | Affected | Fixed |
|---|---|---|
torchPyPI |
before 2.10.0 | 2.10.0 |
pytorchAzure Linux:3 |
before 2.2.2-11 | 2.2.2-11 |
pytorchAzure Linux:2 |
before 2.0.0-14 | 2.0.0-14 |
pytorchBitnami |
before 2.10.0 | 2.10.0 |
pytorchDebian:12 |
All versions, plus listed versions 1.13.1+dfsg-41.13.1+dfsg-52.0.1+dfsg-12.0.1+dfsg-1~exp12.0.1+dfsg-2 Show 33 moreShow less2.0.1+dfsg-42.0.1+dfsg-52.1.2+dfsg-12.1.2+dfsg-22.1.2+dfsg-42.12.0+dfsg2-12.12.0+dfsg2-1~exp12.12.0+dfsg2-1~exp22.12.0+dfsg2-22.12.0+dfsg2-32.12.0+dfsg2-42.12.1+dfsg-12.4.1-12.4.1-32.4.1-42.5.0+dfsg-12.5.1+dfsg-12.5.1+dfsg-32.5.1+dfsg-42.6.0+dfsg-12.6.0+dfsg-1~exp12.6.0+dfsg-22.6.0+dfsg-32.6.0+dfsg-42.6.0+dfsg-52.6.0+dfsg-72.6.0+dfsg-82.6.0+dfsg-92.6.0~rc9+dfsg-1~exp12.9.0+dfsg-1~exp12.9.0+dfsg-1~exp22.9.1+dfsg-1~exp12.9.1+dfsg-1~exp2 |
None known |
pytorchDebian:13 |
All versions, plus listed versions 2.12.0+dfsg2-12.12.0+dfsg2-1~exp12.12.0+dfsg2-1~exp22.12.0+dfsg2-22.12.0+dfsg2-3 Show 9 moreShow less2.12.0+dfsg2-42.12.1+dfsg-12.6.0+dfsg-72.6.0+dfsg-82.6.0+dfsg-92.9.0+dfsg-1~exp12.9.0+dfsg-1~exp22.9.1+dfsg-1~exp12.9.1+dfsg-1~exp2 |
None known |
pytorchDebian:14 |
before 2.12.0+dfsg2-1, plus listed versions 2.12.0+dfsg2-1~exp12.12.0+dfsg2-1~exp22.6.0+dfsg-72.6.0+dfsg-82.6.0+dfsg-9 Show 4 moreShow less2.9.0+dfsg-1~exp12.9.0+dfsg-1~exp22.9.1+dfsg-1~exp12.9.1+dfsg-1~exp2 |
2.12.0+dfsg2-1 |
torchPyPI |
before 2.10.0, plus listed versions 1.0.01.0.11.1.01.10.01.10.1 Show 39 moreShow less1.10.21.11.01.12.01.12.11.13.01.13.11.2.01.3.01.3.11.4.01.5.01.5.11.6.01.7.01.7.11.8.01.8.11.9.01.9.12.0.02.0.12.1.02.1.12.1.22.2.02.2.12.2.22.3.02.3.12.4.02.4.12.5.02.5.12.6.02.7.02.7.12.8.02.9.02.9.1 |
2.10.0 |
pytorchPyPI |
before 2.10.0, plus listed versions 0.1.21.0.2 |
2.10.0 |
pytorchUbuntu:22.04:LTS |
All versions, plus listed versions 1.7.1-71.8.1-4 |
None known |
pytorchUbuntu:25.10 |
All versions, plus listed versions 2.6.0+dfsg-5build12.6.0+dfsg-7 |
None known |
pytorchUbuntu:26.10 |
All versions, plus listed versions 2.9.1+dfsg-1~exp1ubuntu22.12.1+dfsg-1 |
None known |
pytorchUbuntu:26.04:LTS |
All versions, plus listed versions 2.6.0+dfsg-72.6.0+dfsg-92.9.1+dfsg-1~exp1ubuntu2 |
None known |
History
| Recorded | Changed | Sources |
|---|---|---|
| affected, references | OSV | |
| affected, facts, references | GitHub Advisories | |
| first indexed | CVE List, EUVD, GitHub Advisories, NVD, OSV |
Dates are when Every CVE recorded each change. Upstream dates may differ.
In short
CVE-2026-24747 is a high vulnerability (CVSS 8.8) in Linuxfoundation Pytorch, published January 27, 2026. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.80%, the 56th percentile. Fixed in torch 2.10.0.
References
github.com/pytorch/pytorch/security/advisories/GHSA-63cw-57p8-fm3p
github.com/pytorch/pytorch/issues/163105
github.com/pytorch/pytorch/163122/commit/954dc5183ee9205cbe79876ad05dd2d9ae752139
github.com/pytorch/pytorch/releases/tag/v2.10.0
access.redhat.com/security/cve/CVE-2026-24747
bugzilla.redhat.com/show_bug.cgi?id=2433612
Show 12 moreShow less
security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-24747.json
access.redhat.com/errata/RHSA-2026:24977
nvd.nist.gov/vuln/detail/CVE-2026-24747
github.com/advisories/GHSA-63cw-57p8-fm3p
github.com/pypa/advisory-database/tree/main/vulns/pytorch/PYSEC-2026-1856.yaml
github.com/pypa/advisory-database/tree/main/vulns/torch/PYSEC-2026-2286.yaml
github.com/pytorch/pytorch
pypi.org/project/pytorch
github.com/CVEProject/cvelistV5/tree/main/cves/2026/24xxx/CVE-2026-24747.json
security-tracker.debian.org/tracker/CVE-2026-24747
ubuntu.com/security/CVE-2026-24747
cve.org/CVERecord?id=CVE-2026-24747