Paper
Optimizing On-Device Inference for Apple Silicon
Perplexity built Lily, a custom inference engine for Apple Silicon, to run Qwen3.6-35B-A3B more efficiently than general-purpose frameworks like MLX-LM. On a...
Related papers
8
-
Purdue University
-
Rutgers University
Looking for more?
Open the arXiv / PDF link for the full paper. Abstract and authors appear here when enrichment has run.