🐙 GitHub Detail
humanrouter/ddtree-mlx
By humanrouter
Tree-based speculative decoding for Apple Silicon (MLX). ~10-15% faster than DFlash on code, ~1.5x over autoregressive. First MLX port with custom Metal kernels for hybrid model support.
Live Snapshot
⭐
Stars
140
🍴
Forks
11
📄
License
Unknown
🧩
Type
Python
About this open-source project
Live information fetched from GitHub.
Tree-based speculative decoding for Apple Silicon (MLX). ~10-15% faster than DFlash on code, ~1.5x over autoregressive. First MLX port with custom Metal kernels for hybrid model support.
Default Branch
main
Open Issues
0
Watchers
140