Local Ai
b9760
Release b9760 of llama.cpp includes a server refactoring/generalization of the input file schema and wire-up of input_video with raw base64 support. The release includes multiple build variants across
Release b9760 of llama.cpp includes a server refactoring/generalization of the input file schema and wire-up of input_video with raw base64 support. The release includes multiple build variants across platforms including Ubuntu SYCL, macOS Apple Silicon, and others. Released on June 22, 2026, this build represents the ongoing development of llama.cpp, the C/C++ implementation for running large language models efficiently on consumer hardware.
Source: llama.cpp Releases | 2026-06-22