SeferAI searches Sefaria's library — Tanakh, Talmud, Midrash, Halakhah, Kabbalah and commentary — and writes you a synthesis with every claim linked back to the text it came from. Then it can read it aloud.
These are unedited answers from the public library, written by SeferAI for people who asked. No account needed to read them.
Three steps. The first two need no account.
In English, Hebrew, German, Russian, Ukrainian, French, Spanish or Portuguese; the answer comes back in the language you asked in. A verse, a sugya, a theme across sources, or something you have been chewing on all week.
A structured answer, with inline citations that link straight to the primary text on Sefaria. You can check every claim. Choose who it is written for: an essay, scholarly notes, a lesson plan, an answer for beginners, material for a dvar Torah, or a version to read to children.
Turn it into a narrated podcast episode you can play on a walk, or subscribe to in any podcast app. This step needs a free account.
A question is matched against passages, not against page titles. So how many passages there are, and in which language, is most of what determines whether the right source comes back.
| Where the depth went (English and Hebrew) | Texts | Passages |
|---|---|---|
| Commentary — Rishonim, Acharonim, modern | 6,116 | 2,365,241 |
| Halakhah | 383 | 227,592 |
| Talmud | 187 | 204,158 |
| Liturgy | 87 | 196,662 |
| Chasidut | 162 | 170,009 |
| Responsa | 163 | 140,279 |
| Jewish Thought, Kabbalah, Midrash, Musar | 449 | 349,912 |
| Tanakh, Targum, Mishnah, Tosefta, Second Temple | 528 | 148,675 |
The depth is concentrated where the old index was thinnest: commentary. Asking what the Rishonim make of a verse now has six thousand commentary texts behind it rather than a sample of the English ones.
They are separate choices, and neither one is a trap.
You can ask and read without signing in, and it is free. Those questions and their answers join the public study library, where anyone can read them — that library is how most people find this site. Making a podcast is the one thing that needs an account, because it costs real money to narrate.
Sign in with your email — no password, we send you a link — and your syntheses stay private unless you choose to share one. Signing in is free.
SeferAI is not trying to make a profit. Every question costs real money to answer — model time, and narration for the podcasts — and support is what keeps it running and free for people who cannot pay.
Prefer to give once rather than subscribe? A one-off contribution helps just as much.
Curious what the Ultra tier actually produces? Listen to three full episodes — no sign-up, and each comes with the written study behind it.
A language model first restates your question in the words the sources themselves use: the Hebrew terms, their different spellings, and the names of the books where the topic lives. Those words are looked up in a full-text index of about 4.05 million passages of Sefaria's texts, in Hebrew, English, German, French and Portuguese, from Tanakh, the Talmud, Midrash, Halakhah, Kabbalah and commentary. The model then reads the candidates and puts first the ones that discuss your question directly rather than merely mention a word from it. Where the words find too little, a search by meaning fills in.
Yes. Ask in English, Hebrew, German, Russian, Ukrainian, French, Spanish or Portuguese and the answer comes back in the same language, and the whole site can be switched to any of the eight. Sefaria publishes German, French and Portuguese translations of many texts, and those are searched and quoted directly. It has few Spanish ones and almost no Russian or Ukrainian ones, so a Spanish, Russian or Ukrainian answer is written from the Hebrew and English sources and translated, and it says so. Wherever a translation stands between you and the source, the answer tells you.
A frontier large language model reads the retrieved passages and writes the synthesis from them, citing as it goes. It works from the sources it was given rather than from memory, which is what makes the citations checkable — and why you should check them.
You choose, from the Style menu next to Synthesis. Essay (the default) is a written study: the answer in brief, the background, the sources, where they agree and disagree. Scholarly is study notes: the reference first, the key words quoted, disputes stated. Teacher is a lesson plan with the sources in teaching order and questions for the class. Beginner explains every term in plain words. Darshan builds one idea from the sources into a draft dvar Torah. Children is written to be read aloud to children aged 8 to 12, true to the sources and with nothing frightening. Every style draws on the same sources, and none adds a welcome, a sales pitch or a closing thank-you. If your question names one collection (the Talmud, the Mishnah, the Tanakh, the Zohar, the siddur), only that collection is searched, and the answer says so at the top.
Yes. SeferAI is available as a connector for Claude, so you can search the library from inside any Claude conversation and get the references back with links. How to connect it.
Sefaria is the open library of Jewish texts that SeferAI draws from, and it is a separate organisation. Syntheses are study aids produced by a language model. They are not rulings, and they should be attributed to SeferAI rather than to Sefaria or to any authority quoted inside them.