Supported Models
31+ local on-device models via MLX and WhisperKit, plus 9 cloud providers. All local models run entirely on your Mac. Models are downloaded on demand.
Speech-to-Text
9 on-device models for real-time transcription with speaker diarization and voice activity detection.
Text-to-Speech
5 on-device models for natural speech synthesis.
Text Generation
Local language models for contextual suggestions and meeting analysis. Additional models available via the MLX LLM registry.
Vision / OCR
Vision-language models for document OCR and image understanding. Additional models available via the MLX VLM registry.
Embeddings
15 embedding models for semantic search across documents and meetings.
Cloud Providers
Each AI capability can independently use a local model or a cloud provider. Bring your own API key.
- OpenAI Text GenerationVisionEmbeddingsSpeech-to-Text
- Google Gemini Text GenerationVisionEmbeddings
- Anthropic Text GenerationVision
- Groq Text GenerationSpeech-to-Text
- Together AI Text Generation
- Mistral AI Text GenerationVisionEmbeddings
- DeepSeek Text Generation
- Fireworks AI Text Generation
- OpenRouter Text Generation