sefineh-ai/Amharic-Tokenizer
Syllable-aware BPE tokenizer for the Amharic language (አማርኛ) – fast, accurate, trainable.
Decomposes Amharic fidel characters into atomic components before applying BPE merges, then recomposes during detokenization to guarantee perfect round-trip fidelity. Cython-optimized core accelerates the cleaning→decomposition→tokenization→detokenization pipeline. Provides both a CLI tool and Python API with pretrained vocab (30K tokens, `amh_bpe_v0.2.6`) alongside full trainability on custom corpora.
Stars
98
Forks
14
Language
Python
License
MIT
Category
Last pushed
Nov 17, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/sefineh-ai/Amharic-Tokenizer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
eliben/go-sentencepiece
Go implementation of the SentencePiece tokenizer
mdabir1203/BPE_Tokenizer_Visualizer
A Visualizer to check how BPE Tokenizer in an LLM Works
U4RASD/r-bpe
R-BPE: Improving BPE-Tokenizers with Token Reuse
jmaczan/bpe-tokenizer
Byte-Pair Encoding tokenizer for training large language models on huge datasets
franciszekparma/GBPET
GPT-style language model with Byte Pair Encoding tokenizer, built from scratch in PyTorch.