The catalogue lists every Upaniṣad of the Muktikā canon — the hundred and eight that Rāma enumerates to Hanumān in मुक्तिकोपनिषद् 1.30–39 — grouped by Veda, by class (mukhya · sāmānya-vedānta · sannyāsa · yoga · vaiṣṇava · śaiva · śākta), in the canonical order, or as one plain A–Z list. Arrange sets both the catalogue and the text dropdown, so the dropdown can be read either under its Veda headings or as a flat list of names. Click a card to read the text; the अष्टोत्तरशतोपनिषदः title at top left always returns to the catalogue.

Search filters the catalogue by name or Muktikā number as you type. Turn on Search the texts to search the corpus itself instead; inside a text, a bare reference like 2.9 jumps to that mantra. Click the अ button (desktop) to type Devanāgarī directly in compact ITRANS.

Roman spelling is taken loosely. A name or a word typed in roman is matched through a tolerant key, so sannyasa, saṁnyāsa and sanyasa all find संन्यासोपनिषत्, and atman, krishna or upanishad find their words in the texts — aspirates, nasals, ṛ/ri, t/d and doubled letters are all treated as the same. Devanāgarī and Bengali queries are matched exactly.

Script switches the whole page between Devanāgarī, Bengali and roman (IAST). The text is stored once, in Devanāgarī, and transliterated as it is drawn — the Bengali follows Sanskrit orthography (খণ্ড-ত, homorganic anusvāra, য়), not a codepoint swap.

Click a word to see its sandhi-split, grammar and meaning. The gloss comes from the word index built for the Śaṅkara-bhāṣya corpus, so it answers for the vocabulary those texts share — 84.2% of the words here. A word it does not know says so; nothing is guessed.

Translation. Each verse of the canonical 108 can be read with an English and a বাংলা rendering under it — translation switches between both, one, or none. For the ten principal Upaniṣads the translation is the one made for the companion Śaṅkara-bhāṣya reader, carried over verse by verse; it was matched on the text rather than on verse numbers, because the two editions number differently, and only where the two readings agree closely. The rest were translated for this edition under the same terminology glossary, so the whole reads as one edition.

The 120 further texts the collection carries are translated too, with the exceptions named below. That includes the Anubhūtiprakāśa, Vidyāraṇya's verse restatement of what twenty Upaniṣads mean in 2,844 ślokas; the Śākta and Vaiṣṇava tantric manuals; the yoga and saṃnyāsa texts; and the commentaries — Upaniṣad-brahma-yogin's, Ānandavana's, Rāghavendra's — whose lemma-then-gloss shape is kept in the rendering.

Not translated. Eight of the further texts are a canonical Upaniṣad printed again with nothing added — an accented recension, an edition with the anvaya interleaved, an abridgement — and four more carry a commentary this site already holds, translated verse by verse, in its companion Śaṅkara-bhāṣya reader. Nothing in those twelve is absent from the reader; only the presentation differs. Everything else is translated in full.

Commentaries sit under their verses. Fourteen of the further texts print a commentary run together with the text it explains. In those, each verse card carries a commentary chip that unfolds the commentary on that verse, the way the Śaṅkara-bhāṣya reader shows its bhāṣya; show all commentary above the text opens them all. Which passage explains which verse was worked out from the text itself: a passage counts as verse where its words are found in the plain edition of the same Upaniṣad in this collection (or before the edition's own iti … samāptā), and a commentary passage goes under the verse it quotes most, never out of the edition's order. Where the edition fused a verse into its commentary, that passage is left open as a card marked verse with its commentary. A search or a link into a commentary opens it. Three texts keep their commentary inline: the Nīlarudra dīpikā and the Sītā commentary, which give nothing to separate them by, and Gauḍapāda's Kārikā, most of which explains no single mantra. The Daśopaniṣatsāra's explanation is its author's own, part of the one work.

Source of the text: the Upaniṣad collection of sanskritdocuments.org, which is the only complete machine-readable edition of all 108, encoded and proofread by named volunteers over three decades. Each text carries its own encoder, proofreader and revision date at the foot of the page. The ten principal Upaniṣads that Śaṅkara commented on are also served here in full, with the bhāṣya, by the companion reader prasthānatrayī · śāṅkarabhāṣyasahitā (text from Advaitaśāradā, Sringeri Sharada Peetham) — the cards for those texts link across.

Please note: these files are prepared by volunteers for personal study and research. Reposting them elsewhere is asked to go through sanskritdocuments.org first.