/nuketorch.ai

Qwen3.5-9B fine-tuning on four MacBook Pros

Apple’s WWDC26 LoRA fine-tuning recipe, pipelined across four MacBook Pro M5 Max with NukeTorch.

5,358.5

tokens/s · NukeTorch, four MacBook Pro M5 Max

~600

tokens/s · Apple’s WWDC26 run, four M3 Ultra (Apple’s figure)

8×+

faster than Apple’s WWDC26 run, every run

A 4-bit model of 106.3 GB, run across four 48 GB MacBook Pro M5 Max with NukeTorch pipeline parallelism.

177.7

tokens/s · one user

859.0

tokens/s · 32 users at once

NVIDIA’s GR00T N1.7 robot policy, 192 clients at once on four MacBook Pro M5 Max with NukeTorch.

152.07

policy chunks/s, four MacBook Pro M5 Max

2.45×

faster than the fastest published H100 result

4.24×

faster than NVIDIA’s own published H100 figure