TI-2026-26459 CVE-2026-34159
llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
Priority HIGH 82
llama.cpp is an inference of several LLM models in C/C++.
ggml-org
Records53All Threatist records
Exploited (KEV)0None in CISA KEV
CVSS critical14CVSS 9.0–10.0
CVSS high18CVSS 7.0–8.9
Exploited now
ggml-org records in CISA KEVNo ggml-org vulnerabilities are listed in CISA KEV.
Components
Products in the NVD configurations · select one to filter the recordsRecords
29 records in Llama.cpp · all components
Priority HIGH 81
llama.cpp provides LLM inference in C/C++.
Priority HIGH 76
llama.cpp is an inference of several LLM models in C/C++.
Priority HIGH 76
llama.cpp provides LLM inference in C/C++.
TI-2024-15627 CVE-2024-23605 · GHSA-c9jh-pp3m-mqr9
Integer overflow in ggml-org llama.cpp
Priority HIGH 76
A heap-based buffer overflow vulnerability exists in the GGUF library header.n_kv functionality of llama.cpp Commit 18c2e17.
TI-2024-14482 CVE-2024-21802 · GHSA-gwh5-fh5x-ww83
Heap-based buffer overflow in ggml-org llama.cpp
Priority HIGH 76
A heap-based buffer overflow vulnerability exists in the GGUF library info->ne functionality of llama.cpp Commit 18c2e17.
Priority HIGH 73
llama.cpp is an inference of several LLM models in C/C++.
Priority HIGH 73
llama.cpp is an inference of several LLM models in C/C++.
Priority HIGH 73
Llama.cpp is LLM inference in C/C++.
Priority HIGH 72
llama.cpp provides LLM inference in C/C++.
TI-2026-25709 CVE-2026-33298
llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing
Priority HIGH 69
llama.cpp is an inference of several LLM models in C/C++.
TI-2026-22284 CVE-2026-27940
llama.cpp has a Heap Buffer Overflow via Integer Overflow in `mem_size` Calculation — Bypass of CVE-2025-53630 Fix
Priority HIGH 69
llama.cpp is an inference of several LLM models in C/C++.
Priority MEDIUM 62
llama.cpp provides LLM inference in C/C++.
TI-2024-15539 CVE-2024-23496 · GHSA-w4wv-vq5v-xp26
Integer overflow in ggml-org llama.cpp
Priority MEDIUM 46
A heap-based buffer overflow vulnerability exists in the GGUF library gguf_fread_str functionality of llama.cpp Commit 18c2e17.
TI-2024-14511 CVE-2024-21836 · GHSA-fhv7-8956-mv3h
Integer overflow in ggml-org llama.cpp
Priority MEDIUM 46
A heap-based buffer overflow vulnerability exists in the GGUF library header.n_tensors functionality of llama.cpp Commit 18c2e17.
TI-2024-14500 CVE-2024-21825 · GHSA-5998-89hj-xhfc
Integer overflow in ggml-org llama.cpp
Priority MEDIUM 46
A heap-based buffer overflow vulnerability exists in the GGUF library GGUF_TYPE_ARRAY/GGUF_TYPE_STRING parsing functionality of llama.cpp Commit 18c2e17.
TI-2026-31733 CVE-2026-43629 · GHSA-7p8h-8h8j-472c
llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restore
Priority MEDIUM 43
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…
TI-2026-31736 CVE-2026-43632 · GHSA-3f2h-96vw-c84x
llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints
Priority MEDIUM 42
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/coun…
TI-2026-31735 CVE-2026-43631 · GHSA-6hc7-9rph-cm99
llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
Priority MEDIUM 42
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…
TI-2026-37750 CVE-2026-52132 · GHSA-rqxm-jx33-wrqr
llama.cpp through commit 97f06e9, when started with the --reranking flag, allows remote…
Priority MEDIUM 40
llama.cpp through commit 97f06e9, when started with the --reranking flag, allows remote attackers to cause a denial of service (std::bad_alloc and HTTP 500) via a negative top_n value in a POST request to /rerank.
TI-2026-37749 CVE-2026-52131 · GHSA-jfjx-j76r-xrpf
llama.cpp b5693 and before has a Reachable Assertion via the gguf_reader::read function.
Priority LOW 39
llama.cpp b5693 and before has a Reachable Assertion via the gguf_reader::read function.
TI-2026-37748 CVE-2026-52130 · GHSA-8q78-m7w9-j45m
llama.cpp b5693 and before is vulnerable to Uncontrolled Recursion in…
Priority LOW 39
llama.cpp b5693 and before is vulnerable to Uncontrolled Recursion in common/json-schema-to-grammar.cpp, resulting in a denial of service.
TI-2026-49348 CVE-2026-70638 · GHSA-p56f-gv6x-v6mg
llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cpp
Priority LOW 38
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 paramete…
TI-2026-31732 CVE-2026-43628 · GHSA-wvf4-v63v-mwm5
llama.cpp b3978–b9058 Integer Underflow via DRY Sampler
Priority LOW 38
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…
TI-2026-31731 CVE-2026-43627 · GHSA-6gxr-4p8w-gxc9
llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
Priority LOW 38
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.