Local Ai
b9259
Release b9259 of llama.cpp fixes a nullptr crash in the speculative common module's get_devices_str function (PR #23386). The fix addresses an issue where ggml_backend_dev_by_name appends a nullptr se
Release b9259 of llama.cpp fixes a nullptr crash in the speculative common module's get_devices_str function (PR #23386). The fix addresses an issue where ggml_backend_dev_by_name appends a nullptr sentinel to the devices vector, and skipping these nullptr entries prevents assertion failures in ggml_backend_dev_name. This is a bug fix release for the llama.cpp project, which provides C/C++ inference for large language models.
Source: llama.cpp Releases | 2026-05-21