A security flaw has been discovered in ggml-org whisper.cpp up to 1.8.2. This vulnerability affects the function whisper_model_load of the file ggml/src/ggml.c. The manipulation results in null pointer dereference. Attacking locally is a requirement. The exploit has been released to the public and may be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.
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.
CVE-2026-33298 is a high-severity heap-based buffer overflow vulnerability in the ggml-org llama.cpp library, affecting versions prior to b7824. The flaw arises from an integer overflow in the ggml_nbytes function when processing crafted GGUF files with malicious tensor dimensions, causing memory size miscalculations. This leads to a heap buffer overflow during tensor processing, enabling potential remote code execution via memory corruption. Exploitation requires local access and user interaction but no privileges. The vulnerability impacts confidentiality, integrity, and availability of affected systems. A fix is included in version b7824. No known exploits are currently reported in the wild. Organizations using llama.cpp for LLM inference should urgently update to the patched version to mitigate risk.
llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file - CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.
A flaw has been found in ggml-org llama.cpp up to 55abc39. Impacted is the function llama_grammar_advance_stack of the file llama.cpp/src/llama-grammar.cpp of the component GBNF Grammar Handler. This manipulation causes stack-based buffer overflow. The attack needs to be launched locally. The exploit has been published and may be used. Patch name: 18993. To fix this issue, it is recommended to deploy a patch.
llama.cpp is an inference of several LLM models in C/C++. In commits 55d4206c8 and prior, the n_discard parameter is parsed directly from JSON input in the llama.cpp server's completion endpoints without validation to ensure it's non-negative. When a negative value is supplied and the context fills up, llama_memory_seq_rm/add receives a reversed range and negative offset, causing out-of-bounds memory writes in the token evaluation loop. This deterministic memory corruption can crash the process or enable remote code execution (RCE). There is no fix at the time of publication.
CVE-2025-14569 is a use-after-free vulnerability in the read_audio_data function of whisper.cpp versions up to 1.8.2, part of the ggml-org project. The flaw allows local attackers with limited privileges to exploit memory corruption, potentially leading to data corruption or application crashes. No user interaction is required, but local access and low privileges are necessary. The vulnerability has a medium severity score of 4.8 and currently no known exploits in the wild. The vendor has not yet responded to the issue. European organizations using whisper.
llama.cpp is an inference of several LLM models in C/C++. Integer Overflow in the gguf_init_from_file_impl function in ggml/src/gguf.cpp can lead to Heap Out-of-Bounds Read/Write. This vulnerability is fixed in commit 26a48ad699d50b6268900062661bd22f3e792579.
llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs.unsigned integer overflow in llama.cpp's tokenizer implementation (llama_vocab::tokenize) (src/llama-vocab.cpp:3036) resulting in unintended behavior in tokens copying size comparison. Allowing heap-overflowing llama.cpp inferencing engine with carefully manipulated text input during tokenization process. This issue has been patched in version b5721.
llama.cpp is an inference of several LLM models in C/C++. Prior to version b5662, an attacker‐supplied GGUF model vocabulary can trigger a buffer overflow in llama.cpp’s vocabulary‐loading code. Specifically, the helper _try_copy in llama.cpp/src/vocab.cpp: llama_vocab::impl::token_to_piece() casts a very large size_t token length into an int32_t, causing the length check (if (length < (int32_t)size)) to be bypassed. As a result, memcpy is still called with that oversized size, letting a malicious model overwrite memory beyond the intended buffer. This can lead to arbitrary memory corruption and potential code execution. This issue has been patched in version b5662.