ANE Fieldfare

ANE Fieldfare is a tool that enables fast inference of the Gemma 4 26B-A4B large language model on Apple Silicon Macs by leveraging the Apple Neural Engine and GPU for efficient token generation. It includes a command line interface and an OpenAI-compatible HTTP API server allowing integration with chat applications. It optimizes memory use and decoding speed for local AI model execution on compatible Macs running macOS 26 or later.