CVE-2025-3136 json
PyTorch CUDACachingAllocator.cpp torch.cuda.memory.caching_allocator_delete memory corruption
published last yr. modified 2mo ago also BIT-pytorch-2025-3136 , EUVD-2025-9636 , PYSEC-2025-197
- cvss
- 3.3
- version 3.1, CNA
- cisa kev
- Not listed
- not in CISA's catalog
- epss
- 0.3%
- EPSS 0.3% chance of exploitation in 30 days, percentile 19%
- public exploits
- 0
- none known
Summary
A vulnerability, which was classified as problematic, has been found in PyTorch 2.6.0. This issue affects the function torch.cuda.memory.caching_allocator_delete of the file c10/cuda/CUDACachingAllocator.cpp. The manipulation leads to memory corruption. An attack has to be approached locally. The exploit has been disclosed to the public and may be used.
Patch / fix
- bitnami pytorch 2.7.0
Is CVE-2025-3136 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 FencerVendor advisories
What the vendors themselves say about their products: affected, fixed or not affected, and what to do.
red hat product security vex CVE-2025-3136 updated cvss 3.3
PyTorch: PyTorch CUDACachingAllocator.cpp torch.cuda.memory.caching_allocator_delete memory corruption
-
Red Hat All currently supported Red Hat productsnot 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
- last yr.
Exploit prediction
0.27% chance of exploitation in the next 30 days, higher than 19% of all scored CVEs (FIRST EPSS, 2026-11-09).
Exploits and checks
- github.com https://github.com/pytorch/pytorch/issues/149821#issue-2940838975 referenced PoC, tagged "exploit" by a source, unverified
- github.com https://github.com/ARPANET-cybersecurity/vuldb/issues/2 referenced PoC, tagged "exploit" by a source, unverified
- vuldb.com https://vuldb.com/?submit.525252 referenced PoC, tagged "exploit" by a source, unverified
- github.com https://github.com/pytorch/pytorch/issues/149821 referenced PoC, tagged "exploit" by a source, unverified
- github.com https://github.com/pytorch/pytorch/issues/149821#issuecomment-2765311086 referenced PoC, tagged "exploit" by a source, unverified
CVSS scores
| assigned by | version | score | vector |
|---|---|---|---|
| CNA | 4.0 | 4.8 | CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N |
| CNA | 3.1 | 3.3 | CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L |
| CNA | 3.0 | 3.3 | CVSS:3.0/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L |
| CNA | 2.0 | 1.7 | AV:L/AC:L/Au:S/C:N/I:N/A:P |
| CNA | 4.0 | 4.8 | CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X |
Affected packages
| ecosystem | package | affected | fixed in |
|---|---|---|---|
| bitnami | pytorch | >=2.6.0, <2.7.0 |
2.7.0 |
| pypi | torch | <=2.6.0-NA |
no fix listed |
Affected products
- n/a PyTorch
Classification
References
-
github.com
/pytorch/pytorch/issues/149821#issue-2940838975
exploit, issue-tracking, vendor-advisory, advisory, report
-
github.com
/ARPANET-cybersecurity/vuldb/issues/2
exploit, not-applicable, web, report
-
vuldb.com
/?submit.525252
third-party-advisory, exploit, vdb-entry, advisory, evidence
-
github.com
/pytorch/pytorch/issues/149821
issue-tracking, exploit, vendor-advisory, advisory, report
-
github.com
/pytorch/pytorch/issues/149821#issuecomment-2765311086
issue-tracking, exploit, vendor-advisory, advisory, report
-
vuldb.com
/?id.303041
vdb-entry, technical-description, third-party-advisory, advisory
-
vuldb.com
/?ctiid.303041
signature, permissions-required, vdb-entry, report
-
nvd.nist.gov
/vuln/detail/CVE-2025-3136
web
Sources
- CVE List CVE-2025-3136
- NVD CVE-2025-3136
- OSV BIT-pytorch-2025-3136
- OSV PYSEC-2025-197
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.