Sanskrit संस्कृतम्

ISO 639-3: sax I L

183,249 Words in vocabulary

4.44x Best compression

0.8264 Best isotropy

Sample text

Excerpts from Sanskrit Wikipedia articles.

सः यादवकुलस्य राजा आसीत्। प्राचीनवंशावली स्टब्स् प्राप्तः भाषानुबन्धः अपूर्णलेखा...

सः अयोध्याकुलस्य राजा आसीत्। प्राचीन-वंशावली अयोध्याकुल स्टब्स् अपूर्णलेखाः योजन...

स्वर्णगौरीव्रतम् इत्युक्ते गौरीतृतीया एव । तत्र द्रष्टव्यम् । स्टब्स् अपूर्णलेखा...

The 20 most frequently used words in Sanskrit Wikipedia.

Explore Sanskrit interactively with browser-based demos.

Key metrics for all model types at a glance.

from wikilangs import tokenizer
tok = tokenizer('latest', 'sa', 32000)
tokens = tok.tokenize("Your text here")

from wikilangs import ngram
ng = ngram('latest', 'sa', gram_size=3)
score = ng.score("Your text here")

from wikilangs import markov
mc = markov('latest', 'sa', depth=3)
text = mc.generate(length=50)

from wikilangs import vocabulary
vocab = vocabulary('latest', 'sa')
info = vocab.lookup("word")

from wikilangs import embeddings
emb = embeddings('latest', 'sa', dimension=64)
vec = emb.embed_word("word")

Model Type	Variants	Description
Tokenizers	8k, 16k, 32k, 64k	BPE tokenizers with different vocabulary sizes
N-gram (Word)	2, 3, 4, 5-gram	Word-level language models
N-gram (Subword)	2, 3, 4, 5-gram	Subword-level language models
Markov (Word)	Depth 1–5	Word-level text generation
Markov (Subword)	Depth 1–5	Subword-level text generation
Vocabulary	—	Word dictionary with frequency and IDF
Embeddings	32d, 64d, 128d	Position-aware word embeddings