I build the tools I wish existed.
Pinned Loading
-
fast-cuda-cross-attention
fast-cuda-cross-attention PublicCUDA cross-attention kernels for Perciever model with progressive optimization techniques (warp-level, tiling, vectorization) and performance benchmarking scripts.
Python 3
-
Fast-WordPiece-tokenizer
Fast-WordPiece-tokenizer PublicReimplementation of Fast WordPiece Tokenization based on the paper of the same name
Python
-
claudecode-mini-statusbar
claudecode-mini-statusbar PublicTiny monochrome Claude Code status line showing context-window fill, usage limits, and prompt-cache expiry
Python 5
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.

