Sanskrit, with its grammar shown.
Every word of the Bhagavad Gītā and Kālidāsa's Meghadūta, analysed by scholars and checked against Pāṇini's rules. Tap any word to see why it is what it is.
AI that teaches Sanskrit, and Sanskrit that tests AI.
Read classical texts with every word explained
The analysis you see above is checked twice: once by scholars who annotated the corpus, once by a working implementation of the Aṣṭādhyāyī. Where they disagree, we show it.
- LiveReader for the Bhagavad Gītā (700 verses) and the Meghadūta (122 verses): word analysis, sūtra derivations, meter
- NextPaste any verse and get a verified word-by-word reading
- NextSpoken Samskritam practice partner
- NextRecitation coach for meter and pronunciation
A grammar benchmark that no model can talk its way through
Every answer is fixed by Pāṇini's rules or by two independent sources, never by another model. Items and grader are open.
| Task | What the model must do | Items |
|---|---|---|
| Tiṅanta | Produce a verb form from root, lakāra, person and number | 150 |
| Subanta | Decline a stem in a given case and number | 100 |
| Analysis | Identify lakāra, puruṣa and vacana of a verb form | 100 |
| Gītā in context | Give a word's lemma and grammar inside a verse | 100 |
| Chandas | Scan a half-verse into guru and laghu syllables | 50 |
How today's models handle Pāṇini
| Model | Overall | Tiṅanta | Subanta | Analysis | Gītā | Chandas |
|---|
What the two sources say together
Compound members are not yet checked, so they are left out of the agreement figure.
Hear when new texts, voice and the benchmark results arrive.
One email when there is something real to show. No sharing, and you can ask to be removed any time.
Built on open scholarship
Word analyses: the Digital Corpus of Sanskrit by Oliver Hellwig, licensed CC BY 4.0. Its Gītā follows the critical edition of the Mahābhārata; verse numbers here use the common numbering, which differs in chapter 13. Meghadūta verses are numbered as in the corpus, counted separately in the Pūrvamegha and Uttaramegha.
Grammar engine: Vidyut from the Ambuda project, which implements more than 2,000 rules of the Aṣṭādhyāyī.