Critical
PyTorch: `torch.load` with `weights_only=True` leads to remote code execution
Type: advisorySources: OSV
PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.
- Products
- Not specified
- Affected versions
- Not specified
- Fixed versions
- Not specified
- CVEs
- CVE-2025-32434
- Confirmed fixed CVEs
- Not specified
- Updated in this feed
- 3 Nov 2026, 04:06 UTC · Revision 1
Source evidence
PyTorch: `torch.load` with `weights_only=True` leads to remote code execution
core-osv · Original sourcePublished 18 Apr 2025, 15:48 UTC
PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.
osv:CVE-2025-32434
Complete source record
{
"id": "71cc7dd1488984efbc59dcafeeb3e2f6a8db6de8c4d8417a77b2ebac1bd1692c",
"url": "https://osv.dev/vulnerability/CVE-2025-32434",
"method": "structured",
"payload": {
"kind": "advisory",
"title": "PyTorch: `torch.load` with `weights_only=True` leads to remote code execution",
"summary": "PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution (RCE) vulnerability exists in PyTorch when loading a model using torch.load with weights_only=True. This issue has been patched in version 2.6.0.",
"evidence": "osv:CVE-2025-32434",
"products": [],
"severity": "critical",
"structured": {
"id": "CVE-2025-32434",
"aliases": [
"BIT-pytorch-2025-32434",
"CVE-2025-32434",
"GHSA-53q9-r3pm-6pq6",
"PYSEC-2025-41"
],
"metrics": [
{
"type": "CVSS_V4",
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N"
}
],
"affected": [
{
"ranges": [
{
"repo": "https://github.com/pytorch/pytorch",
"type": "GIT",
"events": [
{
"introduced": "0"
},
{
"fixed": "1eba9b3aa3c43f86f4a2c807ac8e12c4a7767340"
}
],
"database_specific": {
"extracted_events": [
{
"range": [
{
"introduced": "0"
},
{
"fixed": "2.6.0"
}
],
"source": "AFFECTED_FIELD",
"strategy": "StringRangeExpression"
},
{
"cpe": "cpe:2.3:a:linuxfoundation:pytorch:*:*:*:*:*:python:*:*",
"range": [
{
"introduced": "0"
},
{
"fixed": "2.6.0"
}
],
"source": "CPE_RANGE"
}
]
}
}
],
"versions": [
"v2.6.0-rc9",
"v2.6.0-rc8",
"v2.6.0-rc7",
"v2.6.0-rc6",
"v2.6.0-rc5",
"v2.6.0-rc3",
"v2.6.0-rc2",
"v2.6.0-rc1",
"cslpull90",
"cslpull89",
"cslpull91",
"cslpull92",
"cslpull88",
"cslpull86",
"cslpull87",
"cslpull81",
"cslpull84",
"cslpull85",
"cslpull76",
"cslpull77",
"cslpull82",
"cslpull83",
"cslpull80",
"cslpull78",
"cslpull79",
"cslpull75",
"forpull1",
"ciflow/unstable/123",
"bc2caa7fdf006894eff7af936babde69ab5a40f8-huydhn-debug",
"ciflow/periodic/3e98831",
"ciflow/slow/da3aba1e46157c4df504b067477cdf2b3c96b194-sdym",
"ciflow/slow/0d5b74da0cab798fbfdb9caa53fad816999c8386-sdym",
"ciflow/slow/8751002215790a3a88750faa8f4366933e296693-sdym",
"ciflow/slow/a837609",
"ciflow/slow/788ff06",
"ciflow/slow/1faef89",
"ciflow/slow/187eb7c",
"ciflow/slow/01c7106",
"ciflow/slow/3920ec1",
"ciflow/slow/0577043",
"ciflow/slow/a206e8b",
"ciflow/slow/59a3759",
"ciflow/slow/1732077",
"ciflow/slow/9d85864",
"ciflow/slow/0e81104",
"ciflow/slow/3b7c6b2",
"ciflow/slow/70ef0bb",
"ciflow/slow/af841f3",
"ciflow/slow/9ffad5b",
"ciflow/periodic/94512-point",
"ciflow/inductor/3d4b92b",
"ciflow/inductor/3b9a386",
"ciflow/inductor/d224ac7",
"ciflow/periodic/csl/test87519",
"ciflow/periodic/csltest88761",
"ciflow/periodic/csltest88275",
"ciflow/periodic/317eeb8",
"ciflow/periodic/2a6d37d",
"ciflow/periodic/3c32",
"ciflow/periodic/054a2fd",
"ciflow/periodic/sha-ec5b83",
"malfet/tag-2ef5611",
"malfet/tag-317b1a0",
"malfet/tag-ec6f767",
"nightly-binary",
"v1.8.0-rc1",
"v1.4.0a0",
"v1.3.0a0",
"v1.2.0a0",
"v1.1.0a0",
"v1.0rc1",
"v1.0rc0",
"v1.0.0a0",
"v0.1.11",
"v0.1.10",
"v0.1.9",
"v0.1.8",
"v0.1.7",
"v0.1.6",
"v0.1.5",
"v0.1.4",
"v0.1.3",
"v0.1.2",
"v0.1.1"
],
"database_specific": {
"source": "https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-32434.json"
}
}
],
"modified": "2026-10-28T02:30:25.879868992Z",
"provider": "osv",
"references": [
"https://lists.debian.org/debian-lts-announce/2025/12/msg00000.html",
"https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/32xxx/CVE-2025-32434.json",
"https://github.com/pytorch/pytorch/security/advisories/GHSA-53q9-r3pm-6pq6",
"https://nvd.nist.gov/vuln/detail/CVE-2025-32434"
],
"provider_metadata": {
"cwe_ids": [
"CWE-502"
],
"cna_assigner": "GitHub_M",
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2025/32xxx/CVE-2025-32434.json"
}
},
"related_cves": [
{
"value": "CVE-2025-32434",
"evidence": "/id"
}
],
"security_relevant": true,
"severity_evidence": "/severity/0/score"
},
"source_id": "core-osv",
"document_id": 492399,
"published_at": "2025-04-18T15:48:18.851+00:00"
}