CVE-2026-73325 Fujitsu OneCompression < 1.2.1 Arbitrary Code Execution via torch.load Deserialization
Attackers can execute arbitrary code by providing a malicious model checkpoint file to the OneCompression library due to unsafe pickle deserialization.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Local or same network only
- Victim action needed
- A user has to click or open something
- Fix
- None listed
What is CVE-2026-73325?
CVE-2026-73325 is a high-severity unsafe deserialization vulnerability in Fujitsu Research OneCompression, published on 12 Aug 2026. It can only be exploited from the same machine or local network, and only if a user opens a file or follows a link.
Official description
Fujitsu Research's OneCompression library before 1.2.1 contains an unsafe deserialization vulnerability that allows attackers to execute arbitrary code by supplying a crafted model.pt checkpoint file, as QuantizedModelLoader.load_quantized_model_pt() unconditionally calls torch.load with weights_only=False, invoking Python's pickle machinery during deserialization. Attackers can embed malicious __reduce__ methods in a crafted model checkpoint to execute arbitrary Python code, including system commands, when the library loads the file from a caller-selected model directory.
Weakness: CWE-502
Which versions are affected by CVE-2026-73325?
How do I fix CVE-2026-73325?
No fix has been published for CVE-2026-73325 yet. Limit exposure of the affected software until one is available.
Upgrade, and only deserialise data from trusted sources or with an allow-list of types. More about Unsafe deserialization.
Is CVE-2026-73325 being exploited?
There is no known exploitation of CVE-2026-73325 in the wild and no public exploit code.
Sources
Also known as GHSA-f3q7-jj9x-2wjr.