dipampaul17/KVSplit
Run larger LLMs with longer contexts on Apple Silicon by using differentiated precision for KV cache quantization. KVSplit enables 8-bit keys & 4-bit values, reducing memory by 59% with <1% quality loss. Includes benchmarking, visualization, and one-command setup. Optimized for M1/M2/M3 Macs with Metal support.
Language: Python
#apple_silicon #generative_ai #kv_cache #llama_cpp #llm #m1 #m2 #m3 #memory_optimization #metal #optimization #quantization
Stars: 222 Issues: 1 Forks: 5
https://github.com/dipampaul17/KVSplit

olvvier/apple-silicon-accelerometer
reading the undocumented mems accelerometer + gyroscope on apple silicon macbooks via iokit hid
Language: Python
#accelerometer #apple #applespu #gyroscope #hid #iokit #m2 #m3 #m4 #macbook #macos #mems #research #sensor #spu
Stars: 551 Issues: 0 Forks: 33
https://github.com/olvvier/apple-silicon-accelerometer