macOS Requirements
RecommendedSystem requirements and GPU acceleration support for macOS.
Apple Silicon (Recommended)
Apple Silicon Macs (M1, M2, M3, M4 and their Pro/Max/Ultra variants) provide the best experience with full Metal GPU acceleration for both transcription and summarization.
M1 / M2 Series
Supported- Full Metal GPU acceleration
- 8GB RAM: Basic models (Whisper Small, Qwen 1.7B)
- 16GB RAM: All models including Large
M3 / M4 Series
Best Performance- Enhanced Metal GPU with dynamic caching
- Faster inference speeds
- Better memory efficiency
Recommended Configurations
| RAM | Whisper Model | LLM Model | Use Case |
|---|---|---|---|
| 8 GB | Whisper Small | Qwen 3 1.7B | Basic usage, English only |
| 16 GB | Whisper Medium | Qwen 3 4B | Recommended for most users |
| 24-32 GB | Whisper Large V3 | Qwen 3 8B | Maximum quality, multilingual |
Intel Macs
Limited Support
Intel Macs do not have GPU acceleration for AI workloads. Transcription and summarization run on CPU only, which is significantly slower than Apple Silicon.
Recommendations for Intel Macs
- Use smaller models: Whisper Tiny/Small and Qwen 1.7B or DeepSeek R1 1.5B
- Increase CPU threads in Settings > Advanced for better parallelization
- Expect longer processing times (3-5x slower than Apple Silicon)
- Consider upgrading to an Apple Silicon Mac for best experience
macOS Version Requirements
Minimum
macOS 12
Monterey
Recommended
macOS 14
Sonoma
Latest
macOS 15
Sequoia