| ▲ | How Apple fits a LLM on iPhone with surgical efficiency(youtube.com) | |
| 9 points by lroe 15 hours ago | 1 comments | ||
| ▲ | paxy 12 hours ago | parent [-] | |
TL;DR - Quantization - KV caching - LoRA - Sparse MoE model - Dedicated AI chip - Offload to cloud Nothing really groundbreaking. | ||