ggml records
29 published records for vendor ggml.
Researcher profile
- Entered KEV
- 0 · 0%
- Weaponized
- 0 · 0%
- Pre-auth RCE
- 5
- With a fix record
- 41.4%
- Median publish → KEV
- No record has entered KEV
Recurring classes
- CWE-190 Integer Overflow or Wraparound6
- CWE-122 Heap-based Buffer Overflow4
- CWE-125 Out-of-bounds Read4
- CWE-119 Improper Restriction of Operations within the Bounds of a Memory Buffer3
- CWE-362 Concurrent Execution using Shared Resource with Improper Synchronization ('Race Condition')2
- CWE-476 NULL Pointer Dereference2
The weakness classes this vendor ships most often: where to look.
CWEAll records
29 records| Action | CVE | Vulnerability | Severity | KEV | EPSS | Published |
|---|---|---|---|---|---|---|
40Plan | CVE-2024-42479No exploit | llama.cpp allows write-what-where in rpc_server::set_tensorggml · llama.cpp · CWE-123 | Critical9.8 | — | 2.6% | Aug 12, 2024 |
39Monitor | CVE-2026-34159Proof of concept | llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backendggml · llama.cpp · CWE-119 | Critical9.8 | — | 1.5% | Apr 1, 2026 |
39Monitor | CVE-2024-21802No exploit | A heap-based buffer overflow vulnerability exists in the GGUF library info->ne functionality of llama.cpp Commit 18c2e17.ggml · llama.cpp · CWE-122 | Critical9.8 | — | 1.3% | Feb 26, 2024 |
39Monitor | CVE-2024-23496No exploit | A heap-based buffer overflow vulnerability exists in the GGUF library gguf_fread_str functionality of llama.cpp Commit 18c2e17.ggml · llama.cpp · CWE-190 | Critical9.8 | — | 1.3% | Feb 26, 2024 |
39Monitor | CVE-2024-23605No exploit | A heap-based buffer overflow vulnerability exists in the GGUF library header.n_kv functionality of llama.cpp Commit 18c2e17.ggml · llama.cpp · CWE-190 | Critical9.8 | — | 1.3% | Feb 26, 2024 |
39Monitor | CVE-2024-21825No exploit | A heap-based buffer overflow vulnerability exists in the GGUF library GGUF_TYPE_ARRAY/GGUF_TYPE_STRING parsing functionality of llama.cpp Coggml · llama.cpp · CWE-190 | Critical9.8 | — | 1.3% | Feb 26, 2024 |
39Monitor | CVE-2024-21836No exploit | A heap-based buffer overflow vulnerability exists in the GGUF library header.n_tensors functionality of llama.cpp Commit 18c2e17.ggml · llama.cpp · CWE-190 | Critical9.8 | — | 1.3% | Feb 26, 2024 |
39Monitor | CVE-2024-42478No exploit | llama.cpp allows Arbitrary Address Read in rpc_server::get_tensorggml · llama.cpp · CWE-125 | Critical9.8 | — | 0.5% | Aug 12, 2024 |
39Monitor | CVE-2026-21869No exploit | llama.cpp has Out-of-bounds Write in llama-serverggml · llama.cpp · CWE-787 | Critical9.8 | — | 0.5% | Jan 7, 2026 |
36Monitor | CVE-2026-43629No exploit | llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restoreggml · llama.cpp · CWE-190 | Critical9.2 | — | 0.6% | Aug 6, 2026 |
36Monitor | CVE-2026-43631No exploit | llama.cpp b7492–b9060 Use-After-Free RCE via llama-serverggml · llama.cpp · CWE-362 | Critical9.2 | — | 0.5% | Aug 6, 2026 |
36Monitor | CVE-2026-43632No exploit | llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpointsggml · llama.cpp · CWE-367 | Critical9.2 | — | 0.4% | Aug 6, 2026 |
35Monitor | CVE-2024-32878No exploit | Use of Uninitialized Variable Vulnerability in llama.cppggml · llama.cpp · CWE-456 | High8.8 | — | 0.6% | Apr 26, 2024 |
35Monitor | CVE-2025-49847No exploit | llama.cpp Vulnerable to Buffer Overflow via Malicious GGUF Modelggml · llama.cpp · CWE-119 | High8.8 | — | 0.5% | Jun 17, 2025 |
35Monitor | CVE-2025-52566No exploit | llama.cpp tokenizer signed vs. unsigned heap overflowggml · llama.cpp · CWE-119 | High8.8 | — | 0.3% | Jun 24, 2025 |
34Monitor | CVE-2026-43628No exploit | llama.cpp b3978–b9058 Integer Underflow via DRY Samplerggml · llama.cpp · CWE-125 | High8.5 | — | 0.2% | Aug 6, 2026 |
34Monitor | CVE-2026-43622No exploit | llama.cpp b1886–b7445 Double Free via llama-android.cppggml · llama.cpp · CWE-415 | High8.5 | — | 0.2% | Aug 6, 2026 |
34Monitor | CVE-2026-70638Proof of concept | llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cppggml · llama.cpp · CWE-122 | High8.5 | — | 0.2% | Aug 6, 2026 |
34Monitor | CVE-2026-43627No exploit | llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Functionggml · llama.cpp · CWE-190 | High8.5 | — | 0.2% | Aug 6, 2026 |
31Monitor | CVE-2026-33298No exploit | llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsingggml · llama.cpp · CWE-122 | High7.8 | — | 0.4% | Mar 23, 2026 |
31Monitor | CVE-2026-27940Proof of concept | llama.cpp has a Heap Buffer Overflow via Integer Overflow in `mem_size` Calculation — Bypass of CVE-2025-53630 Fixggml · llama.cpp · CWE-122 | High7.8 | — | 0.1% | Mar 12, 2026 |
30Monitor | CVE-2026-52132No exploit | llama.cpp through commit 97f06e9, when started with the --reranking flag, allows remote attackers to cause a denial of service (std::bad_allggml · llama.cpp · CWE-674 | High7.5 | — | 0.5% | Sep 1, 2026 |
30Monitor | CVE-2026-52130No exploit | llama.cpp b5693 and before is vulnerable to Uncontrolled Recursion in common/json-schema-to-grammar.cpp, resulting in a denial of service.ggml · llama.cpp · CWE-674 | High7.5 | — | 0.4% | Sep 1, 2026 |
30Monitor | CVE-2024-42477No exploit | llama.cpp global-buffer-overflow in ggml_type_sizeggml · llama.cpp · CWE-125 | High7.5 | — | 0.4% | Aug 12, 2024 |
30Monitor | CVE-2026-52131No exploit | llama.cpp b5693 and before has a Reachable Assertion via the gguf_reader::read function.ggml · llama.cpp · CWE-617 | High7.5 | — | 0.4% | Sep 1, 2026 |
- CVE-2024-4247940Plan
llama.cpp allows write-what-where in rpc_server::set_tensor
CriticalCVSS 9.8No exploitEPSS 2%ggml · llama.cppAug 12, 2024
- CVE-2026-3415939Monitor
llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
CriticalCVSS 9.8Proof of conceptEPSS 1%ggml · llama.cppApr 1, 2026
- CVE-2024-2180239Monitor
A heap-based buffer overflow vulnerability exists in the GGUF library info->ne functionality of llama.cpp Commit 18c2e17.
CriticalCVSS 9.8No exploitEPSS 1%ggml · llama.cppFeb 26, 2024
- CVE-2024-2349639Monitor
A heap-based buffer overflow vulnerability exists in the GGUF library gguf_fread_str functionality of llama.cpp Commit 18c2e17.
CriticalCVSS 9.8No exploitEPSS 1%ggml · llama.cppFeb 26, 2024
- CVE-2024-2360539Monitor
A heap-based buffer overflow vulnerability exists in the GGUF library header.n_kv functionality of llama.cpp Commit 18c2e17.
CriticalCVSS 9.8No exploitEPSS 1%ggml · llama.cppFeb 26, 2024
- CVE-2024-2182539Monitor
A heap-based buffer overflow vulnerability exists in the GGUF library GGUF_TYPE_ARRAY/GGUF_TYPE_STRING parsing functionality of llama.cpp Co
CriticalCVSS 9.8No exploitEPSS 1%ggml · llama.cppFeb 26, 2024
- CVE-2024-2183639Monitor
A heap-based buffer overflow vulnerability exists in the GGUF library header.n_tensors functionality of llama.cpp Commit 18c2e17.
CriticalCVSS 9.8No exploitEPSS 1%ggml · llama.cppFeb 26, 2024
- CVE-2024-4247839Monitor
llama.cpp allows Arbitrary Address Read in rpc_server::get_tensor
CriticalCVSS 9.8No exploitEPSS 0%ggml · llama.cppAug 12, 2024
- CVE-2026-2186939Monitor
llama.cpp has Out-of-bounds Write in llama-server
CriticalCVSS 9.8No exploitEPSS 0%ggml · llama.cppJan 7, 2026
- CVE-2026-4362936Monitor
llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restore
CriticalCVSS 9.2No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-4363136Monitor
llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
CriticalCVSS 9.2No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-4363236Monitor
llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints
CriticalCVSS 9.2No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2024-3287835Monitor
Use of Uninitialized Variable Vulnerability in llama.cpp
HighCVSS 8.8No exploitEPSS 0%ggml · llama.cppApr 26, 2024
- CVE-2025-4984735Monitor
llama.cpp Vulnerable to Buffer Overflow via Malicious GGUF Model
HighCVSS 8.8No exploitEPSS 0%ggml · llama.cppJun 17, 2025
- CVE-2025-5256635Monitor
llama.cpp tokenizer signed vs. unsigned heap overflow
HighCVSS 8.8No exploitEPSS 0%ggml · llama.cppJun 24, 2025
- CVE-2026-4362834Monitor
llama.cpp b3978–b9058 Integer Underflow via DRY Sampler
HighCVSS 8.5No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-4362234Monitor
llama.cpp b1886–b7445 Double Free via llama-android.cpp
HighCVSS 8.5No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-7063834Monitor
llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cpp
HighCVSS 8.5Proof of conceptEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-4362734Monitor
llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
HighCVSS 8.5No exploitEPSS 0%ggml · llama.cppAug 6, 2026
- CVE-2026-3329831Monitor
llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing
HighCVSS 7.8No exploitEPSS 0%ggml · llama.cppMar 23, 2026
- CVE-2026-2794031Monitor
llama.cpp has a Heap Buffer Overflow via Integer Overflow in `mem_size` Calculation — Bypass of CVE-2025-53630 Fix
HighCVSS 7.8Proof of conceptEPSS 0%ggml · llama.cppMar 12, 2026
- CVE-2026-5213230Monitor
llama.cpp through commit 97f06e9, when started with the --reranking flag, allows remote attackers to cause a denial of service (std::bad_all
HighCVSS 7.5No exploitEPSS 0%ggml · llama.cppSep 1, 2026
- CVE-2026-5213030Monitor
llama.cpp b5693 and before is vulnerable to Uncontrolled Recursion in common/json-schema-to-grammar.cpp, resulting in a denial of service.
HighCVSS 7.5No exploitEPSS 0%ggml · llama.cppSep 1, 2026
- CVE-2024-4247730Monitor
llama.cpp global-buffer-overflow in ggml_type_size
HighCVSS 7.5No exploitEPSS 0%ggml · llama.cppAug 12, 2024
- CVE-2026-5213130Monitor
llama.cpp b5693 and before has a Reachable Assertion via the gguf_reader::read function.
HighCVSS 7.5No exploitEPSS 0%ggml · llama.cppSep 1, 2026