CVE-2025-55560

Published Sep 25, 2025

Last updated 8 months ago

CVSS high 7.5

Overview

Description
An issue in pytorch v2.7.0 can lead to a Denial of Service (DoS) when a PyTorch model consists of torch.Tensor.to_sparse() and torch.Tensor.to_dense() and is compiled by Inductor.
Source
cve@mitre.org
NVD status
Analyzed
Products
pytorch

Risk scores

CVSS 3.1

Type
Primary
Base score
7.5
Impact score
3.6
Exploitability score
3.9
Vector string
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Severity
HIGH

Weaknesses

nvd@nist.gov
NVD-CWE-noinfo
134c704f-9b21-4f2e-91b3-4a467353bcc0
CWE-400

Social media

Hype score
Not currently trending

  1. 🚨 PyTorch DoS MELTDOWN — CVE-2025-55560! Bitnami PyTorch 2.7.0 is vulnerable to Denial of Service. Update to v2.7.1+ NOW. If you can’t patch immediately, isolate affected endpoints, add rate limits, and monitor for crashes/resource spikes. 🔍 https://t.co/yVYWphVItF http

    @vulert_official

    15 Oct 2025

    1 Impression

    0 Retweets

    0 Likes

    0 Bookmarks

    0 Replies

    0 Quotes

Configurations

Related CVEs

  1. A vulnerability was identified in PyTorch 2.10.0. The affected element is an unknown function of the component pt2 Loading Handler. The manipulation leads to deserialization. The attack can only be performed from a local environment. The exploit is publicly available and might be used. The project was informed of the problem early through a pull request but has not reacted yet.•CVE-2026-4538
  2. 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.•CVE-2026-24747
  3. An issue was discovered in PyTorch v2.5 and v2.7.1. Omission of profiler.stop() can cause torch.profiler.profile (PythonTracer) to crash or hang during finalization, leading to a Denial of Service (DoS).•CVE-2025-63396
  4. pytorch v2.8.0 was discovered to contain an integer overflow in the component torch.nan_to_num-.long().•CVE-2025-55554
  5. A buffer overflow occurs in pytorch v2.7.0 when a PyTorch model consists of torch.nn.Conv2d, torch.nn.functional.hardshrink, and torch.Tensor.view-torch.mv() and is compiled by Inductor, leading to a Denial of Service (DoS).•CVE-2025-55558

References

Sources include official advisories and independent security research.