alt
Hacker News
raffraffraff
•
today at 8:02 AM
•
0 replies
•
view on HN
So it's not about the model (eg TensorFold/Qwen3.8-27B-MLX-4bit) it's about the inference engine?
alt Hacker News