Search
Find a vulnerability
Search criteria
ⓘ
Use this form to refine search results.
Full-text search supports keyword queries with ranking and filtering.
You can combine vendor, product, and sources to narrow results.
Enable “Apply ordering” to sort by date instead of relevance.
42 vulnerabilities found for llama.cpp by ggml-org
CVE-2026-70640 (GCVE-0-2026-70640)
Vulnerability from nvd – Published: 2026-08-06 15:38 – Updated: 2026-08-06 15:47 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Race Condition Use-After-Free via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.
Severity
CWE
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 7.3,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:H/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "CWE-476 NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:47:17.244Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Race Condition Use-After-Free via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70640",
"datePublished": "2026-08-06T15:38:22.994Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-06T15:47:17.244Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-70639 (GCVE-0-2026-70639)
Vulnerability from nvd – Published: 2026-08-06 15:37 – Updated: 2026-08-06 15:47 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Null Pointer Dereference DoS via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.
Severity
5.5 (Medium)
CWE
- CWE-476 - NULL Pointer Dereference
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 6.8,
"baseSeverity": "MEDIUM",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 5.5,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "CWE-476 NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:47:07.635Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Null Pointer Dereference DoS via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70639",
"datePublished": "2026-08-06T15:37:30.567Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-06T15:47:07.635Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-70638 (GCVE-0-2026-70638)
Vulnerability from nvd – Published: 2026-08-06 15:35 – Updated: 2026-08-07 15:17 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-70638",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T15:17:11.050589Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T15:17:24.736Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-122",
"description": "CWE-122 Heap-based Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:46:53.865Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Integer Overflow via new_1batch() in llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70638",
"datePublished": "2026-08-06T15:35:08.090Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-07T15:17:24.736Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43632 (GCVE-0-2026-43632)
Vulnerability from nvd – Published: 2026-08-06 15:53 – Updated: 2026-08-06 15:53 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints
Summary
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
Severity
8.1 (High)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9060",
"status": "affected",
"version": "b7492",
"versionType": "custom"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-367",
"description": "CWE-367 Time-of-check Time-of-use (TOCTOU) Race Condition",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:53:19.379Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b7492\u2013b9060 Use-After-Free in Tokenization Endpoints",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43632",
"datePublished": "2026-08-06T15:53:19.379Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:53:19.379Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43631 (GCVE-0-2026-43631)
Vulnerability from nvd – Published: 2026-08-06 15:52 – Updated: 2026-08-07 13:50 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
Summary
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
Severity
8.1 (High)
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43631",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T13:49:29.592822Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T13:50:01.743Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9060",
"status": "affected",
"version": "b7492",
"versionType": "custom"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:52:09.915Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b7492\u2013b9060 Use-After-Free RCE via llama-server",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43631",
"datePublished": "2026-08-06T15:52:09.915Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-07T13:50:01.743Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43630 (GCVE-0-2026-43630)
Vulnerability from nvd – Published: 2026-08-06 15:50 – Updated: 2026-08-06 15:50 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b5702–b7653 Out-of-Bounds Read Information Disclosure
Summary
llama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.
Severity
6.5 (Medium)
CWE
Assigner
References
3 references
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7653",
"status": "affected",
"version": "b5702",
"versionType": "custom"
},
{
"status": "affected",
"version": "0.17.1",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 6.3,
"baseSeverity": "MEDIUM",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:N/VA:L/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "LOW",
"vulnConfidentialityImpact": "LOW",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "LOW",
"baseScore": 6.5,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "LOW",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:L",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-125",
"description": "CWE-125 Out-of-bounds Read",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-209",
"description": "CWE-209 Generation of Error Message Containing Sensitive Information",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:50:55.022Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/f93c09e2673090b59802f2552848a1cac70598b0"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b5702\u2013b7653 Out-of-Bounds Read Information Disclosure",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43630",
"datePublished": "2026-08-06T15:50:55.022Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:50:55.022Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43629 (GCVE-0-2026-43629)
Vulnerability from nvd – Published: 2026-08-06 15:48 – Updated: 2026-08-06 15:48 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restore
Summary
llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.
Severity
8.1 (High)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b4882",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.16.1",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-787",
"description": "CWE-787 Out-of-bounds Write",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:48:41.246Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b4882\u2013b9058 Buffer Overflow in KV Cache State Restore",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43629",
"datePublished": "2026-08-06T15:48:41.246Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:48:41.246Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43628 (GCVE-0-2026-43628)
Vulnerability from nvd – Published: 2026-08-06 15:45 – Updated: 2026-08-07 15:20 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b3978–b9058 Integer Underflow via DRY Sampler
Summary
llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43628",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T15:20:25.090606Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T15:20:54.518Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b3978",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.11.0",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-191",
"description": "CWE-191 Integer Underflow (Wrap or Wraparound)",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-125",
"description": "CWE-125 Out-of-bounds Read",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:45:52.293Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b3978\u2013b9058 Integer Underflow via DRY Sampler",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43628",
"datePublished": "2026-08-06T15:45:52.293Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-07T15:20:54.518Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43627 (GCVE-0-2026-43627)
Vulnerability from nvd – Published: 2026-08-06 15:40 – Updated: 2026-08-06 15:43 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
Summary
llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.
Severity
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b1283",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.4.0",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-680",
"description": "CWE-680 Integer Overflow to Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:43:06.461Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1283\u2013b9058 Integer Overflow in llama_batch_init() Function",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43627",
"datePublished": "2026-08-06T15:40:19.714Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:43:06.461Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43622 (GCVE-0-2026-43622)
Vulnerability from nvd – Published: 2026-08-06 15:27 – Updated: 2026-08-06 18:44 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Double Free via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43622",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-06T18:31:47.983642Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T18:44:19.139Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-762",
"description": "CWE-762 Mismatched Memory Management Routines",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-415",
"description": "CWE-415 Double Free",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:46:32.286Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Double Free via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43622",
"datePublished": "2026-08-06T15:27:07.820Z",
"dateReserved": "2026-05-01T18:22:45.640Z",
"dateUpdated": "2026-08-06T18:44:19.139Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-18581 (GCVE-0-2026-18581)
Vulnerability from nvd – Published: 2026-08-03 00:30 – Updated: 2026-08-03 19:47
VLAI
EPSS
VEX
Title
ggml-org llama.cpp Jinja Minja Template parser.cpp assertion
Summary
A vulnerability was determined in ggml-org llama.cpp e15efe0. Affected by this issue is some unknown functionality of the file common/jinja/parser.cpp of the component Jinja Minja Template Parser. Executing a manipulation with the input {{9|9|{ can lead to reachable assertion. The attack requires local access. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet.
Severity
SSVC
Exploitation: poc
Automatable: no
Technical Impact: partial
CISA Coordinator (v2.0.3)
CWE
- CWE-617 - Reachable Assertion
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/385409 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/385409/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-18581 | third-party-advisory |
| https://vuldb.com/submit/799118 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25282 | issue-tracking |
| https://drive.proton.me/urls/R8D8NGZ6Z0#C2cVZYn5z1Xc | exploit |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-18581",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-03T19:47:27.610815Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-03T19:47:33.721Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"references": [
{
"tags": [
"exploit"
],
"url": "https://vuldb.com/submit/799118"
}
],
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"modules": [
"Jinja Minja Template Parser"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A vulnerability was determined in ggml-org llama.cpp e15efe0. Affected by this issue is some unknown functionality of the file common/jinja/parser.cpp of the component Jinja Minja Template Parser. Executing a manipulation with the input {{9|9|{ can lead to reachable assertion. The attack requires local access. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 4.8,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:P",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 3.3,
"baseSeverity": "LOW",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L/E:P/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 3.3,
"baseSeverity": "LOW",
"vectorString": "CVSS:3.0/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L/E:P/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 1.7,
"vectorString": "AV:L/AC:L/Au:S/C:N/I:N/A:P/E:POC/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-617",
"description": "Reachable Assertion",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-03T00:30:08.285Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-385409 | ggml-org llama.cpp Jinja Minja Template parser.cpp assertion",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/385409"
},
{
"name": "VDB-385409 | CTI Indicators (IOB, IOC, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/385409/cti"
},
{
"name": "CVE-2026-18581 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-18581"
},
{
"name": "Submit #799118 | ggml-org llama.cpp confirmed commit e15efe0 CWE-617 (Reachable Assertion)",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/799118"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25282"
},
{
"tags": [
"exploit"
],
"url": "https://drive.proton.me/urls/R8D8NGZ6Z0#C2cVZYn5z1Xc"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-08-02T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-08-02T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-08-02T21:03:57.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp Jinja Minja Template parser.cpp assertion",
"x_generator": [
"VulDB PVTS v202608"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-18581",
"datePublished": "2026-08-03T00:30:08.285Z",
"dateReserved": "2026-08-02T18:58:51.627Z",
"dateUpdated": "2026-08-03T19:47:33.721Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-17501 (GCVE-0-2026-17501)
Vulnerability from nvd – Published: 2026-07-27 00:30 – Updated: 2026-07-29 04:43
VLAI
EPSS
VEX
Title
ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion
Summary
A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes uncontrolled recursion. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance.
Severity
SSVC
Exploitation: none
Automatable: yes
Technical Impact: partial
CISA Coordinator (v2.0.3)
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/383354 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/383354/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-17501 | third-party-advisory |
| https://vuldb.com/submit/798504 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25283 | issue-tracking |
| https://github.com/ggml-org/llama.cpp/pull/25308 | issue-trackingpatch |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-17501",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-07-27T17:05:27.959542Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T20:23:44.187Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"modules": [
"JSON-Schema-to-GBNF Conversion"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "analyst",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes uncontrolled recursion. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 5,
"vectorString": "AV:N/AC:L/Au:N/C:N/I:N/A:P/E:ND/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-674",
"description": "Uncontrolled Recursion",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-404",
"description": "Denial of Service",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-07-29T04:43:05.523Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-383354 | ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/383354"
},
{
"name": "VDB-383354 | CTI Indicators (IOB, IOC, TTP, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/383354/cti"
},
{
"name": "CVE-2026-17501 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-17501"
},
{
"name": "Submit #798504 | ggml-org llama.cpp (confirmed on commit e15efe0 / tag b8635+1) Uncontrolled Recursion",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/798504"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25283"
},
{
"tags": [
"issue-tracking",
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/25308"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-07-26T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-07-26T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-07-29T06:44:46.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion",
"x_generator": [
"VulDB PVTS v202607"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-17501",
"datePublished": "2026-07-27T00:30:11.082Z",
"dateReserved": "2026-07-26T17:33:47.225Z",
"dateUpdated": "2026-07-29T04:43:05.523Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-17500 (GCVE-0-2026-17500)
Vulnerability from nvd – Published: 2026-07-27 00:15 – Updated: 2026-07-27 18:03
VLAI
EPSS
VEX
Title
ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference
Summary
A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0. This affects the function _visit_pattern of the file common/json-schema-to-grammar.cpp. The manipulation results in null pointer dereference. The attack can be launched remotely. The pull request to fix this issue awaits acceptance.
Severity
SSVC
Exploitation: none
Automatable: yes
Technical Impact: partial
CISA Coordinator (v2.0.3)
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/383353 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/383353/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-17500 | third-party-advisory |
| https://vuldb.com/submit/798503 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25284 | issue-tracking |
| https://github.com/ggml-org/llama.cpp/pull/25308 | issue-trackingpatch |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-17500",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-07-27T18:03:09.823962Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T18:03:26.332Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "d006858"
},
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0. This affects the function _visit_pattern of the file common/json-schema-to-grammar.cpp. The manipulation results in null pointer dereference. The attack can be launched remotely. The pull request to fix this issue awaits acceptance."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 5,
"vectorString": "AV:N/AC:L/Au:N/C:N/I:N/A:P/E:ND/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-404",
"description": "Denial of Service",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T00:15:12.222Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-383353 | ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/383353"
},
{
"name": "VDB-383353 | CTI Indicators (IOB, IOC, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/383353/cti"
},
{
"name": "CVE-2026-17500 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-17500"
},
{
"name": "Submit #798503 | ggml-org llama.cpp confirmed on commit d006858 / 2025-04-03 and commit e15efe0 / tag b8635+1) Reachable Assertion",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/798503"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25284"
},
{
"tags": [
"issue-tracking",
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/25308"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-07-26T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-07-26T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-07-26T19:38:55.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference",
"x_generator": [
"VulDB PVTS v202607"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-17500",
"datePublished": "2026-07-27T00:15:12.222Z",
"dateReserved": "2026-07-26T17:33:41.700Z",
"dateUpdated": "2026-07-27T18:03:26.332Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-34159 (GCVE-0-2026-34159)
Vulnerability from nvd – Published: 2026-04-01 16:59 – Updated: 2026-04-02 03:56
VLAI
EPSS
VEX
Title
llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
Summary
llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492.
Severity
9.8 (Critical)
SSVC
Exploitation: poc
Automatable: yes
Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
- CWE-119 - Improper Restriction of Operations within the Bounds of a Memory Buffer
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/security/ad… | x_refsource_CONFIRM |
| https://github.com/ggml-org/llama.cpp/pull/20908 | x_refsource_MISC |
| https://github.com/ggml-org/llama.cpp/commit/39bf… | x_refsource_MISC |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-34159",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-04-01T00:00:00+00:00",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-04-02T03:56:11.820Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "\u003c b8492"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend\u0027s deserialize_tensor() skips all bounds validation when a tensor\u0027s buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 9.8,
"baseSeverity": "CRITICAL",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-119",
"description": "CWE-119: Improper Restriction of Operations within the Bounds of a Memory Buffer",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-04-01T16:59:59.967Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw"
},
{
"name": "https://github.com/ggml-org/llama.cpp/pull/20908",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/20908"
},
{
"name": "https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042"
}
],
"source": {
"advisory": "GHSA-j8rj-fmpv-wcxw",
"discovery": "UNKNOWN"
},
"title": "llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-34159",
"datePublished": "2026-04-01T16:59:59.967Z",
"dateReserved": "2026-03-25T20:12:04.197Z",
"dateUpdated": "2026-04-02T03:56:11.820Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-33298 (GCVE-0-2026-33298)
Vulnerability from nvd – Published: 2026-03-24 00:01 – Updated: 2026-03-25 03:55
VLAI
EPSS
VEX
Title
llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing
Summary
llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes `ggml_nbytes` to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix.
Severity
7.8 (High)
SSVC
Exploitation: poc
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
2 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/security/ad… | x_refsource_CONFIRM |
| https://github.com/ggml-org/llama.cpp/releases/ta… | x_refsource_MISC |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-33298",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-03-24T00:00:00+00:00",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-03-25T03:55:51.679Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "\u003c b7824"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes `ggml_nbytes` to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-122",
"description": "CWE-122: Heap-based Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190: Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-03-24T00:01:40.989Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-96jg-mvhq-q7q7",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-96jg-mvhq-q7q7"
},
{
"name": "https://github.com/ggml-org/llama.cpp/releases/tag/b7824",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7824"
}
],
"source": {
"advisory": "GHSA-96jg-mvhq-q7q7",
"discovery": "UNKNOWN"
},
"title": "llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-33298",
"datePublished": "2026-03-24T00:01:40.989Z",
"dateReserved": "2026-03-18T18:55:47.427Z",
"dateUpdated": "2026-03-25T03:55:51.679Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43632 (GCVE-0-2026-43632)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:53 – Updated: 2026-08-06 15:53 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints
Summary
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.
Severity
8.1 (High)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9060",
"status": "affected",
"version": "b7492",
"versionType": "custom"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-367",
"description": "CWE-367 Time-of-check Time-of-use (TOCTOU) Race Condition",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:53:19.379Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b7492\u2013b9060 Use-After-Free in Tokenization Endpoints",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43632",
"datePublished": "2026-08-06T15:53:19.379Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:53:19.379Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43631 (GCVE-0-2026-43631)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:52 – Updated: 2026-08-07 13:50 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
Summary
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
Severity
8.1 (High)
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43631",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T13:49:29.592822Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T13:50:01.743Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9060",
"status": "affected",
"version": "b7492",
"versionType": "custom"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-416",
"description": "CWE-416 Use After Free",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:52:09.915Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b7492\u2013b9060 Use-After-Free RCE via llama-server",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43631",
"datePublished": "2026-08-06T15:52:09.915Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-07T13:50:01.743Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43630 (GCVE-0-2026-43630)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:50 – Updated: 2026-08-06 15:50 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b5702–b7653 Out-of-Bounds Read Information Disclosure
Summary
llama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.
Severity
6.5 (Medium)
CWE
Assigner
References
3 references
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7653",
"status": "affected",
"version": "b5702",
"versionType": "custom"
},
{
"status": "affected",
"version": "0.17.1",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b5702 through b7653 contain an out-of-bounds read vulnerability in the recurrent memory state restore path that allows attackers with write access to the slot save directory to read memory past the end of the allocated cells array. Attackers can craft a malicious slot file with an oversized seq_id value to trigger an out-of-bounds read that leaks heap data including pointer values into server logs, defeating ASLR protections and facilitating further exploitation."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 6.3,
"baseSeverity": "MEDIUM",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:N/VA:L/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "LOW",
"vulnConfidentialityImpact": "LOW",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "LOW",
"baseScore": 6.5,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "LOW",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:L",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-125",
"description": "CWE-125 Out-of-bounds Read",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-209",
"description": "CWE-209 Generation of Error Message Containing Sensitive Information",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:50:55.022Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/f93c09e2673090b59802f2552848a1cac70598b0"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b5702\u2013b7653 Out-of-Bounds Read Information Disclosure",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43630",
"datePublished": "2026-08-06T15:50:55.022Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:50:55.022Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43629 (GCVE-0-2026-43629)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:48 – Updated: 2026-08-06 15:48 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restore
Summary
llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.
Severity
8.1 (High)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b4882",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.16.1",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restore path where the state_read_data() function computes write size without overflow checking, allowing attackers with write access to the slot_save_path directory to corrupt heap memory. Attackers can craft malicious state files where cell_count multiplication overflows or exceeds tensor buffer allocation to write attacker-controlled bytes past buffer boundaries, potentially resulting in heap metadata corruption, model weight corruption, or arbitrary code execution via function pointer overwrite."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "PRESENT",
"attackVector": "NETWORK",
"baseScore": 9.2,
"baseSeverity": "CRITICAL",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "NONE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 8.1,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-787",
"description": "CWE-787 Out-of-bounds Write",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:48:41.246Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b4882\u2013b9058 Buffer Overflow in KV Cache State Restore",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43629",
"datePublished": "2026-08-06T15:48:41.246Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:48:41.246Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43628 (GCVE-0-2026-43628)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:45 – Updated: 2026-08-07 15:20 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b3978–b9058 Integer Underflow via DRY Sampler
Summary
llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43628",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T15:20:25.090606Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T15:20:54.518Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b3978",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.11.0",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-191",
"description": "CWE-191 Integer Underflow (Wrap or Wraparound)",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-125",
"description": "CWE-125 Out-of-bounds Read",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:45:52.293Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b3978\u2013b9058 Integer Underflow via DRY Sampler",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43628",
"datePublished": "2026-08-06T15:45:52.293Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-07T15:20:54.518Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43627 (GCVE-0-2026-43627)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:40 – Updated: 2026-08-06 15:43 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
Summary
llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.
Severity
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/Vladimir-tokarev-cyera/llama-c… | mitigation |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "affected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b9058",
"status": "affected",
"version": "b1283",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.4.0",
"versionType": "semver"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-680",
"description": "CWE-680 Integer Overflow to Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:43:06.461Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"mitigation"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1283\u2013b9058 Integer Overflow in llama_batch_init() Function",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43627",
"datePublished": "2026-08-06T15:40:19.714Z",
"dateReserved": "2026-05-01T18:22:45.641Z",
"dateUpdated": "2026-08-06T15:43:06.461Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-70640 (GCVE-0-2026-70640)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:38 – Updated: 2026-08-06 15:47 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Race Condition Use-After-Free via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.
Severity
CWE
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "HIGH",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 7.3,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:H/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "CWE-476 NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-362",
"description": "CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization (\u0027Race Condition\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:47:17.244Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Race Condition Use-After-Free via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70640",
"datePublished": "2026-08-06T15:38:22.994Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-06T15:47:17.244Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-70639 (GCVE-0-2026-70639)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:37 – Updated: 2026-08-06 15:47 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Null Pointer Dereference DoS via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.
Severity
5.5 (Medium)
CWE
- CWE-476 - NULL Pointer Dereference
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 6.8,
"baseSeverity": "MEDIUM",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "NONE",
"vulnIntegrityImpact": "NONE",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 5.5,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "NONE",
"integrityImpact": "NONE",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "CWE-476 NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:47:07.635Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Null Pointer Dereference DoS via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70639",
"datePublished": "2026-08-06T15:37:30.567Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-06T15:47:07.635Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-70638 (GCVE-0-2026-70638)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:35 – Updated: 2026-08-07 15:17 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-70638",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-07T15:17:11.050589Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-07T15:17:24.736Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190 Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-122",
"description": "CWE-122 Heap-based Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:46:53.865Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Integer Overflow via new_1batch() in llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-70638",
"datePublished": "2026-08-06T15:35:08.090Z",
"dateReserved": "2026-08-04T20:17:18.298Z",
"dateUpdated": "2026-08-07T15:17:24.736Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-43622 (GCVE-0-2026-43622)
Vulnerability from cvelistv5 – Published: 2026-08-06 15:27 – Updated: 2026-08-06 18:44 X_Open Source
VLAI
EPSS
VEX
Title
llama.cpp b1886–b7445 Double Free via llama-android.cpp
Summary
llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.
Severity
SSVC
Exploitation: none
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/releases/ta… | release-notes |
| https://github.com/ggml-org/llama.cpp/commit/5c0d… | patch |
| https://github.com/Vladimir-tokarev-cyera/llama-c… | related |
Impacted products
Date Public
2026-05-06 00:00
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-43622",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-06T18:31:47.983642Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T18:44:19.139Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"defaultStatus": "unaffected",
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"lessThanOrEqual": "b7445",
"status": "affected",
"version": "b1886",
"versionType": "custom"
},
{
"lessThanOrEqual": "0.17.1",
"status": "affected",
"version": "0.9.0",
"versionType": "semver"
},
{
"status": "unaffected",
"version": "5c0d18881e0e9794c96b2602736b758bac9d9388",
"versionType": "git"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "finder",
"value": "Vladimir Tokarev (@G1ND1L4) - Vulnerability Research Tech Lead, Cyera"
},
{
"lang": "en",
"type": "finder",
"value": "Ofek Itach (@ofekitach) - Security Research Team Lead, Cyera"
}
],
"datePublic": "2026-05-06T00:00:00.000Z",
"descriptions": [
{
"lang": "en",
"supportingMedia": [
{
"base64": false,
"type": "text/html",
"value": "\u003cp\u003ellama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.\u003c/p\u003e"
}
],
"value": "llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state."
}
],
"metrics": [
{
"cvssV4_0": {
"Automatable": "NOT_DEFINED",
"Recovery": "NOT_DEFINED",
"Safety": "NOT_DEFINED",
"attackComplexity": "LOW",
"attackRequirements": "NONE",
"attackVector": "LOCAL",
"baseScore": 8.5,
"baseSeverity": "HIGH",
"exploitMaturity": "NOT_DEFINED",
"privilegesRequired": "NONE",
"providerUrgency": "NOT_DEFINED",
"subAvailabilityImpact": "NONE",
"subConfidentialityImpact": "NONE",
"subIntegrityImpact": "NONE",
"userInteraction": "PASSIVE",
"valueDensity": "NOT_DEFINED",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"version": "4.0",
"vulnAvailabilityImpact": "HIGH",
"vulnConfidentialityImpact": "HIGH",
"vulnIntegrityImpact": "HIGH",
"vulnerabilityResponseEffort": "NOT_DEFINED"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
},
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
},
"format": "CVSS",
"scenarios": [
{
"lang": "en",
"value": "GENERAL"
}
]
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-762",
"description": "CWE-762 Mismatched Memory Management Routines",
"lang": "en",
"type": "CWE"
},
{
"cweId": "CWE-415",
"description": "CWE-415 Double Free",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-06T15:46:32.286Z",
"orgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"shortName": "VulnCheck"
},
"references": [
{
"tags": [
"release-notes"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7446"
},
{
"tags": [
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388"
},
{
"tags": [
"related"
],
"url": "https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches"
}
],
"source": {
"discovery": "EXTERNAL"
},
"tags": [
"x_open-source"
],
"title": "llama.cpp b1886\u2013b7445 Double Free via llama-android.cpp",
"x_generator": {
"engine": "vulncheck"
}
}
},
"cveMetadata": {
"assignerOrgId": "83251b91-4cc7-4094-a5c7-464a1b83ea10",
"assignerShortName": "VulnCheck",
"cveId": "CVE-2026-43622",
"datePublished": "2026-08-06T15:27:07.820Z",
"dateReserved": "2026-05-01T18:22:45.640Z",
"dateUpdated": "2026-08-06T18:44:19.139Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-18581 (GCVE-0-2026-18581)
Vulnerability from cvelistv5 – Published: 2026-08-03 00:30 – Updated: 2026-08-03 19:47
VLAI
EPSS
VEX
Title
ggml-org llama.cpp Jinja Minja Template parser.cpp assertion
Summary
A vulnerability was determined in ggml-org llama.cpp e15efe0. Affected by this issue is some unknown functionality of the file common/jinja/parser.cpp of the component Jinja Minja Template Parser. Executing a manipulation with the input {{9|9|{ can lead to reachable assertion. The attack requires local access. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet.
Severity
SSVC
Exploitation: poc
Automatable: no
Technical Impact: partial
CISA Coordinator (v2.0.3)
CWE
- CWE-617 - Reachable Assertion
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/385409 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/385409/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-18581 | third-party-advisory |
| https://vuldb.com/submit/799118 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25282 | issue-tracking |
| https://drive.proton.me/urls/R8D8NGZ6Z0#C2cVZYn5z1Xc | exploit |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-18581",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-08-03T19:47:27.610815Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-08-03T19:47:33.721Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"references": [
{
"tags": [
"exploit"
],
"url": "https://vuldb.com/submit/799118"
}
],
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"modules": [
"Jinja Minja Template Parser"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A vulnerability was determined in ggml-org llama.cpp e15efe0. Affected by this issue is some unknown functionality of the file common/jinja/parser.cpp of the component Jinja Minja Template Parser. Executing a manipulation with the input {{9|9|{ can lead to reachable assertion. The attack requires local access. The exploit has been publicly disclosed and may be utilized. The project was informed of the problem early through an issue report but has not responded yet."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 4.8,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:P",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 3.3,
"baseSeverity": "LOW",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L/E:P/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 3.3,
"baseSeverity": "LOW",
"vectorString": "CVSS:3.0/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L/E:P/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 1.7,
"vectorString": "AV:L/AC:L/Au:S/C:N/I:N/A:P/E:POC/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-617",
"description": "Reachable Assertion",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-08-03T00:30:08.285Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-385409 | ggml-org llama.cpp Jinja Minja Template parser.cpp assertion",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/385409"
},
{
"name": "VDB-385409 | CTI Indicators (IOB, IOC, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/385409/cti"
},
{
"name": "CVE-2026-18581 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-18581"
},
{
"name": "Submit #799118 | ggml-org llama.cpp confirmed commit e15efe0 CWE-617 (Reachable Assertion)",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/799118"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25282"
},
{
"tags": [
"exploit"
],
"url": "https://drive.proton.me/urls/R8D8NGZ6Z0#C2cVZYn5z1Xc"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-08-02T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-08-02T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-08-02T21:03:57.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp Jinja Minja Template parser.cpp assertion",
"x_generator": [
"VulDB PVTS v202608"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-18581",
"datePublished": "2026-08-03T00:30:08.285Z",
"dateReserved": "2026-08-02T18:58:51.627Z",
"dateUpdated": "2026-08-03T19:47:33.721Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-17501 (GCVE-0-2026-17501)
Vulnerability from cvelistv5 – Published: 2026-07-27 00:30 – Updated: 2026-07-29 04:43
VLAI
EPSS
VEX
Title
ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion
Summary
A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes uncontrolled recursion. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance.
Severity
SSVC
Exploitation: none
Automatable: yes
Technical Impact: partial
CISA Coordinator (v2.0.3)
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/383354 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/383354/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-17501 | third-party-advisory |
| https://vuldb.com/submit/798504 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25283 | issue-tracking |
| https://github.com/ggml-org/llama.cpp/pull/25308 | issue-trackingpatch |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-17501",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-07-27T17:05:27.959542Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T20:23:44.187Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"modules": [
"JSON-Schema-to-GBNF Conversion"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "analyst",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes uncontrolled recursion. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 5,
"vectorString": "AV:N/AC:L/Au:N/C:N/I:N/A:P/E:ND/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-674",
"description": "Uncontrolled Recursion",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-404",
"description": "Denial of Service",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-07-29T04:43:05.523Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-383354 | ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/383354"
},
{
"name": "VDB-383354 | CTI Indicators (IOB, IOC, TTP, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/383354/cti"
},
{
"name": "CVE-2026-17501 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-17501"
},
{
"name": "Submit #798504 | ggml-org llama.cpp (confirmed on commit e15efe0 / tag b8635+1) Uncontrolled Recursion",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/798504"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25283"
},
{
"tags": [
"issue-tracking",
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/25308"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-07-26T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-07-26T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-07-29T06:44:46.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp JSON-Schema-to-GBNF Conversion json-schema-to-grammar.cpp transform recursion",
"x_generator": [
"VulDB PVTS v202607"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-17501",
"datePublished": "2026-07-27T00:30:11.082Z",
"dateReserved": "2026-07-26T17:33:47.225Z",
"dateUpdated": "2026-07-29T04:43:05.523Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-17500 (GCVE-0-2026-17500)
Vulnerability from cvelistv5 – Published: 2026-07-27 00:15 – Updated: 2026-07-27 18:03
VLAI
EPSS
VEX
Title
ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference
Summary
A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0. This affects the function _visit_pattern of the file common/json-schema-to-grammar.cpp. The manipulation results in null pointer dereference. The attack can be launched remotely. The pull request to fix this issue awaits acceptance.
Severity
SSVC
Exploitation: none
Automatable: yes
Technical Impact: partial
CISA Coordinator (v2.0.3)
Assigner
References
7 references
| URL | Tags |
|---|---|
| https://vuldb.com/vuln/383353 | vdb-entrytechnical-description |
| https://vuldb.com/vuln/383353/cti | signaturepermissions-required |
| https://vuldb.com/cve/CVE-2026-17500 | third-party-advisory |
| https://vuldb.com/submit/798503 | third-party-advisory |
| https://github.com/ggml-org/llama.cpp/issues/25284 | issue-tracking |
| https://github.com/ggml-org/llama.cpp/pull/25308 | issue-trackingpatch |
| https://github.com/ggml-org/llama.cpp/ | product |
Impacted products
Credits
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-17500",
"options": [
{
"Exploitation": "none"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-07-27T18:03:09.823962Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T18:03:26.332Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"cpes": [
"cpe:2.3:a:ggml-org:llama.cpp:*:*:*:*:*:*:*:*"
],
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "d006858"
},
{
"status": "affected",
"version": "e15efe0"
}
]
}
],
"credits": [
{
"lang": "en",
"type": "reporter",
"value": "m00dy (VulDB User)"
},
{
"lang": "en",
"type": "coordinator",
"value": "VulDB CNA Team"
}
],
"descriptions": [
{
"lang": "en",
"value": "A vulnerability was detected in ggml-org llama.cpp d006858/e15efe0. This affects the function _visit_pattern of the file common/json-schema-to-grammar.cpp. The manipulation results in null pointer dereference. The attack can be launched remotely. The pull request to fix this issue awaits acceptance."
}
],
"metrics": [
{
"cvssV4_0": {
"baseScore": 6.9,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X",
"version": "4.0"
}
},
{
"cvssV3_1": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.1"
}
},
{
"cvssV3_0": {
"baseScore": 5.3,
"baseSeverity": "MEDIUM",
"vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:X/RL:X/RC:R",
"version": "3.0"
}
},
{
"cvssV2_0": {
"baseScore": 5,
"vectorString": "AV:N/AC:L/Au:N/C:N/I:N/A:P/E:ND/RL:ND/RC:UR",
"version": "2.0"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-476",
"description": "NULL Pointer Dereference",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-404",
"description": "Denial of Service",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-07-27T00:15:12.222Z",
"orgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"shortName": "VulDB"
},
"references": [
{
"name": "VDB-383353 | ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference",
"tags": [
"vdb-entry",
"technical-description"
],
"url": "https://vuldb.com/vuln/383353"
},
{
"name": "VDB-383353 | CTI Indicators (IOB, IOC, IOA)",
"tags": [
"signature",
"permissions-required"
],
"url": "https://vuldb.com/vuln/383353/cti"
},
{
"name": "CVE-2026-17500 | CVE Analysis and Report",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/cve/CVE-2026-17500"
},
{
"name": "Submit #798503 | ggml-org llama.cpp confirmed on commit d006858 / 2025-04-03 and commit e15efe0 / tag b8635+1) Reachable Assertion",
"tags": [
"third-party-advisory"
],
"url": "https://vuldb.com/submit/798503"
},
{
"tags": [
"issue-tracking"
],
"url": "https://github.com/ggml-org/llama.cpp/issues/25284"
},
{
"tags": [
"issue-tracking",
"patch"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/25308"
},
{
"tags": [
"product"
],
"url": "https://github.com/ggml-org/llama.cpp/"
}
],
"timeline": [
{
"lang": "en",
"time": "2026-07-26T00:00:00.000Z",
"value": "Advisory disclosed"
},
{
"lang": "en",
"time": "2026-07-26T02:00:00.000Z",
"value": "VulDB entry created"
},
{
"lang": "en",
"time": "2026-07-26T19:38:55.000Z",
"value": "VulDB entry last update"
}
],
"title": "ggml-org llama.cpp json-schema-to-grammar.cpp _visit_pattern null pointer dereference",
"x_generator": [
"VulDB PVTS v202607"
]
}
},
"cveMetadata": {
"assignerOrgId": "1af790b2-7ee1-4545-860a-a788eba489b5",
"assignerShortName": "VulDB",
"cveId": "CVE-2026-17500",
"datePublished": "2026-07-27T00:15:12.222Z",
"dateReserved": "2026-07-26T17:33:41.700Z",
"dateUpdated": "2026-07-27T18:03:26.332Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-34159 (GCVE-0-2026-34159)
Vulnerability from cvelistv5 – Published: 2026-04-01 16:59 – Updated: 2026-04-02 03:56
VLAI
EPSS
VEX
Title
llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
Summary
llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492.
Severity
9.8 (Critical)
SSVC
Exploitation: poc
Automatable: yes
Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
- CWE-119 - Improper Restriction of Operations within the Bounds of a Memory Buffer
Assigner
References
3 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/security/ad… | x_refsource_CONFIRM |
| https://github.com/ggml-org/llama.cpp/pull/20908 | x_refsource_MISC |
| https://github.com/ggml-org/llama.cpp/commit/39bf… | x_refsource_MISC |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-34159",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "yes"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-04-01T00:00:00+00:00",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-04-02T03:56:11.820Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "\u003c b8492"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend\u0027s deserialize_tensor() skips all bounds validation when a tensor\u0027s buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 9.8,
"baseSeverity": "CRITICAL",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-119",
"description": "CWE-119: Improper Restriction of Operations within the Bounds of a Memory Buffer",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-04-01T16:59:59.967Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw"
},
{
"name": "https://github.com/ggml-org/llama.cpp/pull/20908",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/pull/20908"
},
{
"name": "https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042"
}
],
"source": {
"advisory": "GHSA-j8rj-fmpv-wcxw",
"discovery": "UNKNOWN"
},
"title": "llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-34159",
"datePublished": "2026-04-01T16:59:59.967Z",
"dateReserved": "2026-03-25T20:12:04.197Z",
"dateUpdated": "2026-04-02T03:56:11.820Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
CVE-2026-33298 (GCVE-0-2026-33298)
Vulnerability from cvelistv5 – Published: 2026-03-24 00:01 – Updated: 2026-03-25 03:55
VLAI
EPSS
VEX
Title
llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing
Summary
llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes `ggml_nbytes` to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix.
Severity
7.8 (High)
SSVC
Exploitation: poc
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
Assigner
References
2 references
| URL | Tags |
|---|---|
| https://github.com/ggml-org/llama.cpp/security/ad… | x_refsource_CONFIRM |
| https://github.com/ggml-org/llama.cpp/releases/ta… | x_refsource_MISC |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-33298",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-03-24T00:00:00+00:00",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-03-25T03:55:51.679Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "llama.cpp",
"vendor": "ggml-org",
"versions": [
{
"status": "affected",
"version": "\u003c b7824"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes `ggml_nbytes` to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "LOCAL",
"availabilityImpact": "HIGH",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "NONE",
"scope": "UNCHANGED",
"userInteraction": "REQUIRED",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-122",
"description": "CWE-122: Heap-based Buffer Overflow",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-190",
"description": "CWE-190: Integer Overflow or Wraparound",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-03-24T00:01:40.989Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-96jg-mvhq-q7q7",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-96jg-mvhq-q7q7"
},
{
"name": "https://github.com/ggml-org/llama.cpp/releases/tag/b7824",
"tags": [
"x_refsource_MISC"
],
"url": "https://github.com/ggml-org/llama.cpp/releases/tag/b7824"
}
],
"source": {
"advisory": "GHSA-96jg-mvhq-q7q7",
"discovery": "UNKNOWN"
},
"title": "llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-33298",
"datePublished": "2026-03-24T00:01:40.989Z",
"dateReserved": "2026-03-18T18:55:47.427Z",
"dateUpdated": "2026-03-25T03:55:51.679Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}