CVE-2025-52566

llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs. unsigned integer overflow in llama.cpp's tokenizer implementation (llama_vocab::tokenize) (src/llama-vocab.cpp:3036) resulting in unintended behavior in tokens copying size comparison. Allowing heap-overflowing llama.cpp inferencing engine with carefully manipulated text input during tokenization process. This issue has been patched in version b5721.

CVSS v3 8.6 HIGH

8.6^/10

CVSS v3 : HIGH

Vector :

Exploitability : 1.8 / Impact : 6.0

Attack Vector LOCAL

Attack Complexity LOW

Privileges Required NONE

User Interaction REQUIRED

Confidentiality Impact HIGH

Integrity Impact HIGH

Availability Impact HIGH

Scope CHANGED

References

Link	Resource
https://github.com/ggml-org/llama.cpp/commit/dd6e6d0b6a4bbe3ebfc931d1eb14db2f2b1d70af	Patch
https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx	Exploit Vendor Advisory
https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx	Exploit Vendor Advisory

Configurations

Configuration 1 (hide)

cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*

History

27 Aug 2025, 14:01

Type	Values Removed	Values Added
CPE		cpe:2.3:a:ggml:llama.cpp::::::::
First Time		Ggml Ggml llama.cpp
References	~~() https://github.com/ggml-org/llama.cpp/commit/dd6e6d0b6a4bbe3ebfc931d1eb14db2f2b1d70af -~~	() https://github.com/ggml-org/llama.cpp/commit/dd6e6d0b6a4bbe3ebfc931d1eb14db2f2b1d70af - Patch
References	~~() https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx -~~	() https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx - Exploit, Vendor Advisory

26 Jun 2025, 18:58

Type	Values Removed	Values Added
New CVE

Information

Published : 2025-06-24 04:15

Updated : 2025-08-27 14:01

NVD link : CVE-2025-52566

Mitre link : CVE-2025-52566

CVE.ORG link : CVE-2025-52566

JSON object : View

Products Affected

ggml

llama.cpp

CWE

CWE-119

Improper Restriction of Operations within the Bounds of a Memory Buffer

CWE-195

Signed to Unsigned Conversion Error

{"id": "CVE-2025-52566", "cveTags": [], "metrics": {"cvssMetricV31": [{"type": "Secondary", "source": "security-advisories@github.com", "cvssData": {"scope": "CHANGED", "version": "3.1", "baseScore": 8.6, "attackVector": "LOCAL", "baseSeverity": "HIGH", "vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H", "integrityImpact": "HIGH", "userInteraction": "REQUIRED", "attackComplexity": "LOW", "availabilityImpact": "HIGH", "privilegesRequired": "NONE", "confidentialityImpact": "HIGH"}, "impactScore": 6.0, "exploitabilityScore": 1.8}, {"type": "Primary", "source": "nvd@nist.gov", "cvssData": {"scope": "UNCHANGED", "version": "3.1", "baseScore": 8.8, "attackVector": "NETWORK", "baseSeverity": "HIGH", "vectorString": "CVSS:3.1/AV:N/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": 2.8}]}, "published": "2025-06-24T04:15:46.967", "references": [{"url": "https://github.com/ggml-org/llama.cpp/commit/dd6e6d0b6a4bbe3ebfc931d1eb14db2f2b1d70af", "tags": ["Patch"], "source": "security-advisories@github.com"}, {"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx", "tags": ["Exploit", "Vendor Advisory"], "source": "security-advisories@github.com"}, {"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx", "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-119"}, {"lang": "en", "value": "CWE-195"}]}], "descriptions": [{"lang": "en", "value": "llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs. unsigned integer overflow in llama.cpp's tokenizer implementation (llama_vocab::tokenize) (src/llama-vocab.cpp:3036) resulting in unintended behavior in tokens copying size comparison. Allowing heap-overflowing llama.cpp inferencing engine with carefully manipulated text input during tokenization process. This issue has been patched in version b5721."}, {"lang": "es", "value": "llama.cpp es una inferencia de varios modelos LLM en C/C++. Antes de la versi\u00f3n b5721, se produc\u00eda un desbordamiento de enteros con signo y sin signo en la implementaci\u00f3n del tokenizador de llama.cpp (llama_vocab::tokenize) (src/llama-vocab.cpp:3036), lo que provocaba un comportamiento no deseado al comparar el tama\u00f1o de copia de tokens. Esto permit\u00eda el desbordamiento del mont\u00f3n del motor de inferencia de llama.cpp con una entrada de texto cuidadosamente manipulada durante el proceso de tokenizaci\u00f3n. Este problema se ha corregido en la versi\u00f3n b5721."}], "lastModified": "2025-08-27T14:01:31.297", "configurations": [{"nodes": [{"negate": false, "cpeMatch": [{"criteria": "cpe:2.3:a:ggml:llama.cpp:*:*:*:*:*:*:*:*", "vulnerable": true, "matchCriteriaId": "DC465FEB-FFD7-42D5-8D81-F416C28985BD", "versionEndExcluding": "b5721"}], "operator": "OR"}]}], "sourceIdentifier": "security-advisories@github.com"}