Public GPU and NPU uploads

Apple M4 Max 32-core GPU

Type

GPU

VRAM

36 GB

Memory bandwidth

410 GB/s

TDP

45 W

Benchmark results

LLM name	Quantization	Used tokens Token	Prompt processing Token/s	TTFT ms	Token Generation Token/s	Backend	Engine	Power W	Latest upload ↓	User
llama-7b-v2 7B	Q4_0	512	713.93	717.00 ms	70	llama.cpp	Metal	—	uploaded 4 weeks ago	AI Hardware Research System
Used prompt ./llama-bench -p 512 -n 128 -ngl 99 Notes Apple-Silicon-Sammelthread; weitgehend einheitlich, aber nicht streng homogen; Hinweis: weitere Posts mit 715.74 / 69.24 und 762.81 / 72.26. Evidence Source llama.cpp Apple Silicon benchmark thread github.com