Authenticated code execution via unsafe checkpoint deserialization
Published Jun 26, 2025 · Updated Jun 26, 2025
Code injection in hiyouga LlamaFactory 0.9.3 and earlier allows remote authenticated users to execute arbitrary code through its WebUI. During reward-model training, load_valuehead_params retrieves value_head.bin from the supplied checkpoint repository and passes it to torch.load without weights_only=True. Exploitation requires WebUI access and an attacker-controlled checkpoint path, and code runs with the LlamaFactory server process's privileges.
Summary
What happened
Code injection in hiyouga LlamaFactory 0.9.3 and earlier allows remote authenticated users to execute arbitrary code through its WebUI. During reward-model training, load_valuehead_params retrieves value_head.bin from the supplied checkpoint repository and passes it to torch.load without weights_only=True. Exploitation requires WebUI access and an attacker-controlled checkpoint path, and code runs with the LlamaFactory server process's privileges.
The record
- CVE
- CVE-2025-53002
- Published
- Jun 26, 2025
- Updated
- Jun 26, 2025
- Vendor
- hiyouga
- Product
- LlamaFactory
- Classifications
- CWE-94, CWE-502, T1190
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Jun 26, 2025CVE publishedPublication date reported by the CVE source.
- Jun 26, 2025Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 0.9.4
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- GHSA-xj56-p8mm-qmxj WebUI proof of conceptfunctional · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo