- Nativ allows running open models from Google, Cohere, and Liquid AI locally on Apple Silicon Macs, with model recommendations tailored to hardware.
- Features include a clean interface with streaming, markdown, code highlighting, image input, and performance monitoring like tokens/sec and thermal state.
- Built on MLX-VLM and optimized for M-series unified memory and Metal, with no wrappers or translation layers.
- Supports various tasks: chat with LLM, image captioning, video summarization, code autocompletion, and speech transcription/generation.
- Emphasizes privacy: no accounts, subscriptions, credits, or data selling; fully end-to-end owned by users.
- Open-source and MIT licensed, with full transparency in code, model loaders, and telemetry, encouraging community forking and contributions.
- Integrates with existing coding agents by providing local endpoints, maintaining familiar workflows without paywalls or proprietary restrictions.