Threat database

CVE-2026-110810 is a low-severity vulnerability in ggml-org whisper.cpp up to version 1.9.4. It involves an out-of-bounds read in the whisper_model_load function triggered by manipulating the argument ttype. The vulnerability requires local access and low privileges to exploit. A public exploit exists. Upgrading to version 1.9.5 mitigates the issue.

A use-after-free and double free vulnerability exists in llama.cpp before commit b11393 in the common_chat_peg_mapper::map function. This flaw allows unauthenticated remote attackers to corrupt heap memory by submitting a crafted POST /completion request with a dangling current_tool pointer. Exploitation can cause the llama-server to crash and potentially enable heap memory manipulation.

CVE-2026-86317 is a medium severity vulnerability in ggml-org llama.cpp up to version 0.4.0. It affects the RPC Server component, specifically the rpc_server::deserialize_tensor function. The vulnerability arises from a reachable assertion triggered by manipulation of the argument 'ne'. This flaw can be exploited remotely without authentication or user interaction.

CVE-2026-78148 is a medium severity vulnerability in the ggml-org llama.cpp project. It involves a null pointer dereference in the rpc_server::graph_compute function within the ggml-RPC Server component. The flaw can be triggered remotely without authentication or user interaction. A fix has been proposed but is not yet merged or officially released.

CVE-2026-78147 is a medium severity deserialization vulnerability in the ggml-org llama.cpp project. It affects the deserialize_tensor function in the ggml-RPC Server component. The flaw allows remote attackers to manipulate the op/op_params arguments during deserialization. This vulnerability is distinct from a prior issue (CVE-2026-34159) and was reported but closed due to inactivity. No patch or remediation information is currently available.

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.

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.

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.

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.

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.

llama.cpp builds b4882 before b8487 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.

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.

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.

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.

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.

CVE-2026-18581 is a medium severity vulnerability in the ggml-org llama.cpp project affecting the Jinja Minja Template Parser component. A reachable assertion can be triggered by manipulating input with the string {{9|9|{, requiring local access to exploit. The vulnerability has been publicly disclosed, but no official response or patch has been provided by the vendor yet.

A vulnerability was found in ggml-org whisper.cpp 95ea8f9b. Affected is the function ggml_ftype_to_ggml_type of the file ggml/src/ggml.c. The manipulation of the argument ftype results in reachable assertion. The attack requires a local approach. The project was informed of the problem early through an issue report but has not responded yet.

A vulnerability has been found in ggml-org whisper.cpp 1.8.4-58. This impacts the function log_mel_spectrogram of the file src/whisper.cpp. The manipulation leads to out-of-bounds read. The attack needs to be performed locally. The pull request to fix this issue awaits acceptance.

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.

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.