GH Repository · RunanywhereAI
wally
Talk to your Mac, query your docs, no cloud required. On-device voice AI + RAG
- stars
- 1,545
- 30-day movement
- +21/day
- Related entries
- 66
- Connections
- 4
C++qwen3kokoro-ttslocal-ailfm2apple-siliconvoice-assistantllama-cppllmmetalparakeetai-assistantspeech-to-textkitten-ttson-device-aitool-callingragtext-to-speech
wally is a Mac voice assistant that runs entirely on-device, combining speech-to-text, an LLM, text-to-speech, and RAG so you can talk to it and query your own documents without any cloud. It is written in C++, targets Apple Silicon (Metal, llama.cpp), and supports tool calling with models like Qwen3, LFM2, Parakeet, and Kokoro/Kitten TTS.
Reach for it when you want a private, offline voice AI on your Mac that can answer questions from your local docs.
Use it to
- Talk to your Mac for hands-free voice queries
- Run RAG searches over your local documents
- Keep all AI processing fully offline and private
- Use tool calling with local LLMs on Apple Silicon
For Mac users with Apple Silicon wanting private local voice AI
- Role
- rag
- Language
- C++
- Licence
- MIT
- Forks
- 86
- Open issues
- 25
- Last push
- 2026-09-17
- Latest release
- v0.1.2 · 2026-03-04
- Skills shipped
- 6
topicson-device-aivoice-assistantragapple-siliconspeech-to-textlocal-llm