Free Quran and Hadith translation downloads

Quran translations

Every entry covers all 114 surahs (6,236 ayahs). Arabic ships the Uthmani mushaf from quran.com in full. The modern translations (Saheeh International, Kuliev, Musayev, Diyanet) are returned in full text too — each row carries the translator attribution and the upstream source URL so the original author is always credited.

Hadith collections

Nine collections, four languages where coverage exists. Arabic editions are public-domain classical text. Non-Arabic translations split into i-muslim authored (CC0) and upstream mirrored (metadata only) — the response envelope tells you how many of each. Counts below come from the live Firestore tally.

How to use the data

Every endpoint returns a single JSON object. For non-Arabic Hadith endpoints the envelope carries two licence stanzas under sources.authored and sources.imported, plus a per-item source tag so you can filter(i => i.source === "authored") to keep only the CC0-licensed rows. Authored items return full text; upstream-mirrored items return text: null — fetch from source_url if you need them.

# discover everything
curl https://i-muslim.com/api/v1/translations

# full text — Arabic Quran (public domain)
curl https://i-muslim.com/api/v1/translations/quran/ar

# full text — Saheeh International (English Quran)
curl https://i-muslim.com/api/v1/translations/quran/en

# CC0 (i-muslim authored) + metadata (upstream) — mixed envelope
curl https://i-muslim.com/api/v1/translations/hadith/bukhari/en \
  | jq '.data.sources, (.data.items | map(select(.source == "authored")) | length)'

# full text — Bukhari Arabic (classical text)
curl https://i-muslim.com/api/v1/translations/hadith/bukhari/ar

# single surah for lighter payloads
curl https://i-muslim.com/api/v1/translations/quran/ar/1

Need a richer surface (per-ayah, per-hadith, with translations as query parameters, plus write access for your own translations)? See the Developer API docs — those endpoints require an API key but cover the same data at a finer grain.

Licence model

There is no single licence covering the whole dataset. Each item is governed by the licence of its original author, and the API response carries that licence verbatim. Three buckets:

• Full text (CC0) — translations authored by i-muslim (flagged per-item with editedTranslations[lang] in our store). Released under CC0 1.0: do anything, no attribution required.
• Full text — Arabic originals (the Uthmani Quran mushaf and the classical Arabic Hadith editions), public-domain text returned verbatim; and the modern Quran translations (Saheeh International, Kuliev, Musayev, Diyanet), returned in full under their original translator licence with attribution and the upstream source_url preserved on every row.
• Metadata only — non-Arabic Hadith items we haven't re-authored, mirrored from upstream sources like fawazahmed0/hadith-api. The API returns the attribution, licence label and upstream URL but text is null. Fetch the text from source_url.

We default to metadata onlywhenever a third-party translation's redistribution licence is unclear. If you hold the copyright on one of those translations and want to release it under an open licence, get in touch — we'd love to.

Contribute or correct a translation

Spotted an error? Want to contribute a translation you wrote yourself and own? Send it through the contact form with subject “Translation contribution”— include the diff, the source you took the text from, and the licence you're releasing your contribution under. We'll review and merge into the CC0-authored pool.

Open contribution form →

Versioning

These downloads sit under /api/v1/ — the stable surface. Any breaking change to the JSON shape ships under v2; we'll keep v1 running for at least 12 months after. Adding a new translation, flipping a row from metadata-only to full text, or growing the authored-CC0 count is not a breaking change.