slavko-at-klincov-it/Fiber-Inference

Systematic evaluation of all 5 Apple M4 compute units for LLM inference. ANE prefill at 21,490 tok/s, AMX 1.8x faster than GPU — but MLX wins 2.2x. 200+ measurements, full research paper.

22
/ 100
Experimental
No Package No Dependents
Maintenance 13 / 25
Adoption 0 / 25
Maturity 9 / 25
Community 0 / 25

How are scores calculated?

Stars

Forks

Language

Objective-C

License

MIT

Last pushed

Apr 05, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/slavko-at-klincov-it/Fiber-Inference"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.