CVE-2025-46567

LLama Factory enables fine-tuning of large language models. Prior to version 1.0.0, a critical vulnerability exists in the `llamafy_baichuan2.py` script of the LLaMA-Factory project. The script performs insecure deserialization using `torch.load()` on user-supplied `.bin` files from an input directory. An attacker can exploit this behavior by crafting a malicious `.bin` file that executes arbitrary commands during deserialization. This issue has been patched in version 1.0.0.

CVSS v3 6.1 MEDIUM

6.1^/10

CVSS v3 : MEDIUM

Vector :

Exploitability : 1.3 / Impact : 4.7

Attack Vector LOCAL

Attack Complexity LOW

Privileges Required LOW

User Interaction REQUIRED

Confidentiality Impact HIGH

Integrity Impact LOW

Availability Impact LOW

Scope UNCHANGED

References

Link	Resource
https://github.com/hiyouga/LLaMA-Factory/commit/2989d39239d2f46e584c1e1180ba46b9768afb2a	Patch
https://github.com/hiyouga/LLaMA-Factory/security/advisories/GHSA-f2f7-gj54-6vpv	Exploit Vendor Advisory
https://github.com/hiyouga/LLaMA-Factory/security/advisories/GHSA-f2f7-gj54-6vpv	Exploit Vendor Advisory

Configurations

Configuration 1 (hide)

cpe:2.3:a:hiyouga:llama-factory:*:*:*:*:*:*:*:*

History

17 Jun 2025, 14:19

Type	Values Removed	Values Added
New CVE

Information

Published : 2025-05-01 18:15

Updated : 2025-06-17 14:19

NVD link : CVE-2025-46567

Mitre link : CVE-2025-46567

CVE.ORG link : CVE-2025-46567

JSON object : View

Products Affected

hiyouga

llama-factory

CWE

CWE-502

Deserialization of Untrusted Data

{"id": "CVE-2025-46567", "cveTags": [], "metrics": {"cvssMetricV31": [{"type": "Secondary", "source": "security-advisories@github.com", "cvssData": {"scope": "UNCHANGED", "version": "3.1", "baseScore": 6.1, "attackVector": "LOCAL", "baseSeverity": "MEDIUM", "vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:U/C:H/I:L/A:L", "integrityImpact": "LOW", "userInteraction": "REQUIRED", "attackComplexity": "LOW", "availabilityImpact": "LOW", "privilegesRequired": "LOW", "confidentialityImpact": "HIGH"}, "impactScore": 4.7, "exploitabilityScore": 1.3}, {"type": "Primary", "source": "nvd@nist.gov", "cvssData": {"scope": "UNCHANGED", "version": "3.1", "baseScore": 7.8, "attackVector": "LOCAL", "baseSeverity": "HIGH", "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H", "integrityImpact": "HIGH", "userInteraction": "REQUIRED", "attackComplexity": "LOW", "availabilityImpact": "HIGH", "privilegesRequired": "NONE", "confidentialityImpact": "HIGH"}, "impactScore": 5.9, "exploitabilityScore": 1.8}]}, "published": "2025-05-01T18:15:58.117", "references": [{"url": "https://github.com/hiyouga/LLaMA-Factory/commit/2989d39239d2f46e584c1e1180ba46b9768afb2a", "tags": ["Patch"], "source": "security-advisories@github.com"}, {"url": "https://github.com/hiyouga/LLaMA-Factory/security/advisories/GHSA-f2f7-gj54-6vpv", "tags": ["Exploit", "Vendor Advisory"], "source": "security-advisories@github.com"}, {"url": "https://github.com/hiyouga/LLaMA-Factory/security/advisories/GHSA-f2f7-gj54-6vpv", "tags": ["Exploit", "Vendor Advisory"], "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0"}], "vulnStatus": "Analyzed", "weaknesses": [{"type": "Secondary", "source": "security-advisories@github.com", "description": [{"lang": "en", "value": "CWE-502"}]}], "descriptions": [{"lang": "en", "value": "LLama Factory enables fine-tuning of large language models. Prior to version 1.0.0, a critical vulnerability exists in the `llamafy_baichuan2.py` script of the LLaMA-Factory project. The script performs insecure deserialization using `torch.load()` on user-supplied `.bin` files from an input directory. An attacker can exploit this behavior by crafting a malicious `.bin` file that executes arbitrary commands during deserialization. This issue has been patched in version 1.0.0."}, {"lang": "es", "value": "LLama Factory permite el ajuste fino de modelos de lenguaje grandes. Antes de la versi\u00f3n 1.0.0, exist\u00eda una vulnerabilidad cr\u00edtica en el script `llamafy_baichuan2.py` del proyecto LLaMA-Factory. Este script realiza una deserializaci\u00f3n insegura mediante `torch.load()` en archivos `.bin` proporcionados por el usuario desde un directorio de entrada. Un atacante puede explotar este comportamiento manipulando un archivo `.bin` malicioso que ejecute comandos arbitrarios durante la deserializaci\u00f3n. Este problema se ha corregido en la versi\u00f3n 1.0.0."}], "lastModified": "2025-06-17T14:19:39.290", "configurations": [{"nodes": [{"negate": false, "cpeMatch": [{"criteria": "cpe:2.3:a:hiyouga:llama-factory:*:*:*:*:*:*:*:*", "vulnerable": true, "matchCriteriaId": "CCE62620-4488-49A9-BA7B-C8E2790768AC", "versionEndExcluding": "1.0.0"}], "operator": "OR"}]}], "sourceIdentifier": "security-advisories@github.com"}