Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports…
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
davila7
cli
free
Others in the same category, ranked by how often they are opened.
