Ensure your system is ready for real-time AI inference.
| Component | Minimum | Recommended |
|---|---|---|
| Operating System | macOS 13+ (Apple Silicon or Intel) | macOS 14.5+ (Apple Silicon) |
| Processor | Apple M1 / Intel Core i7 (10th Gen+) | Apple M2 / M3 / M4 / M5 |
| Memory | 16 GB RAM (Unified / Shared Memory) | 24 GB+ RAM (Unified Shared Memory) |
| Graphics | Integrated GPU (Metal 3, 6 GB+ shared memory) | Dedicated GPU / Apple Neural Engine (8 GB+ VRAM / Shared Memory) |
Note on VRAM & Unified Shared Memory:
PairHint executes local AI LLM models and speech recognition directly on your system's GPU / NPU. Rather than basic system RAM alone, real-time AI inference requires 6 GB+ of VRAM (on dedicated GPUs) or Unified Shared Memory (on Apple Silicon M1/M2/M3/M4/M5) to hold active model weights in memory for instant responses.