Pinned Loading
-
self-draft
self-draft PublicZero-download MTP self-speculative decoding for llama.cpp: one flag, ~1.5x lower-latency LLM inference on Arm64 cloud, no separate draft model.
Python 1
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.