CVE-2026-43627 llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
Attackers can trigger an integer overflow in llama.cpp to cause heap corruption and achieve arbitrary code execution.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Local or same network only
- Victim action needed
- A user has to click or open something
- Fix
- None listed
What is CVE-2026-43627?
CVE-2026-43627 is a high-severity vulnerability in ggml-org llama.cpp, published on 6 Aug 2026. It can only be exploited from the same machine or local network, and only if a user opens a file or follows a link.
Official description
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.
Which versions are affected by CVE-2026-43627?
It affects 1 product.
Products include Ggml Llama.cpp.
How do I fix CVE-2026-43627?
No fix has been published for CVE-2026-43627 yet. Limit exposure of the affected software until one is available.
Is CVE-2026-43627 being exploited?
There is no known exploitation of CVE-2026-43627 in the wild and no public exploit code.
Other vulnerabilities in ggml-org llama.cpp
- CVE-2024-42479 llama.cpp allows write-what-where in rpc_server::set_tensor
- CVE-2026-34159 llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
- CVE-2026-43629 llama.cpp b4882–b9058 Buffer Overflow in KV Cache State Restore
- CVE-2026-43631 llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
- CVE-2026-43632 llama.cpp b7492–b9060 Use-After-Free in Tokenization Endpoints
Sources
Also known as DEBIAN-CVE-2026-43627, GHSA-6gxr-4p8w-gxc9, UBUNTU-CVE-2026-43627.