CVE-2025-3000 json
PyTorch torch.jit.script memory corruption
published 2025-03-31 19 mo ago modified 2026-09-10 1 mo ago also BIT-pytorch-2025-3000 , EUVD-2025-8759 , GHSA-rrmf-rvhw-rf47 , PYSEC-2025-194
- cvss
- 5.3
- 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 10%
- public exploits
- 0
- none known
Summary
A vulnerability classified as critical has been found in PyTorch 2.6.0. This affects the function torch.jit.script. The manipulation leads to memory corruption. It is possible to launch the attack on the local host. The exploit has been disclosed to the public and may be used.
Patch / fix
CISA SSVC
- Exploitation
- none
- Automatable
- no
- Technical impact
- partial
- Assessed
- 19 mo ago
Exploit prediction
0.20% chance of exploitation in the next 30 days, higher than 10% of all scored CVEs (FIRST EPSS, 2026-11-08).
Exploits and checks
- github.com https://github.com/pytorch/pytorch/issues/149623#issue-2935703015 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:L/VI:L/VA:L/SC:N/SI:N/SA:N |
| CNA | 3.1 | 5.3 | CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L |
| CNA | 3.0 | 5.3 | CVSS:3.0/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L |
| CNA | 2.0 | 4.3 | AV:L/AC:L/Au:S/C:P/I:P/A:P |
| GitHub | 3.1 | 5.3 | CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L |
| CNA | 4.0 | 4.8 | CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/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 |
|---|---|---|---|
| pypi | torch | <2.13.0 <=2.6.0-NA |
2.13.0 |
| bitnami | pytorch | >=2.6.0, <2.7.0 |
2.7.0 |
Affected products
- n/a PyTorch
Classification
- Weakness
-
- CWE-119 Improper Restriction of Operations within the Bounds of a Memory Buffer
References
-
github.com
/pytorch/pytorch/commit/b90c94991cdf8b87c8f7439f79518e0ef2c4ca4f
web
-
github.com
/pytorch/pytorch/issues/149623#issue-2935703015
exploit, issue-tracking, web, vendor-advisory, advisory, report
-
github.com
/pytorch/pytorch/issues/149623
issue-tracking, web, vendor-advisory, advisory, report
-
vuldb.com
/?id.302049
vdb-entry, technical-description, web, third-party-advisory, advisory
-
vuldb.com
/?ctiid.302049
signature, permissions-required, web, vdb-entry, report
-
vuldb.com
/?submit.524197
third-party-advisory, web, vdb-entry, advisory
-
nvd.nist.gov
/vuln/detail/CVE-2025-3000
advisory, web
-
github.com
/pypa/advisory-database/tree/main/vulns/torch/PYSEC-2025-194.yaml
web
-
github.com
/pytorch/pytorch
package
-
github.com
/advisories/GHSA-rrmf-rvhw-rf47
advisory
Sources
- CVE List CVE-2025-3000
- GitHub Advisory GHSA-rrmf-rvhw-rf47
- NVD CVE-2025-3000
- OSV BIT-pytorch-2025-3000
- OSV PYSEC-2025-194
Fencer Vulns 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.