Jina Reader

license

Apache-2.0(OSI)

source

public(TypeScript, stateless mode self-hostable via Docker)

core-gated

ungated(no enterprise directory or license key

managed-tier

r.jina.ai/s.jina.ai hosted SaaS (free tier + paid API key) on top of the same OSS core

Apache-2.0 throughout with no enterprise directory and no license key, and the paid tiers buy throughput on the hosted endpoint rather than functionality: jina.ai/reader lists Reader at 20 RPM without a key, 500 RPM with a free key and 5,000 RPM on Premium, with PDF, MS-Office, image captioning and search available at every tier. The published branch does have the MongoDB-backed SaaS storage layer stripped out, which the README states plainly, but it is accounting and persistence for the hosted service rather than a piece of the reader the open build needs - the README says the branch "runs in stateless mode out of the box, with optional MinIO/S3-compatible bucket caching". Nothing is withheld from the published source, so core-gated is ungated.

• https://github.com/jina-ai/reader recorded 2026-06-04

  Apache-2.0 license, public source, self-host Docker instructions, ~11k stars

• https://github.com/jina-ai/reader/blob/main/LICENSE recorded 2026-08-12

  the repository LICENSE is the Apache License, Version 2.0, headed "Copyright 2020-2024 Jina AI Limited"; the GitHub repo API reports the SPDX id Apache-2.0

• https://github.com/jina-ai/reader/blob/main/README.md recorded 2026-08-12

  the repo is "the open source branch of the codebase behind https://r.jina.ai and https://s.jina.ai"; it "runs in stateless or bucket-cached mode; the MongoDB-backed SaaS storage layer is not included here"; the 2026-04 entry adds that the oss branch "runs in stateless mode out of the box, with optional MinIO/S3-compatible bucket caching via docker compose"

• https://jina.ai/reader recorded 2026-08-12

  the pricing and rate-limit table: Reader 20 RPM with no key, 500 RPM with a free API key, 5,000 RPM on Premium, and Search 100/1,000 RPM; the feature list (PDF reading, image captioning, web search) is the same at every tier, so the tiers differ on throughput rather than on features. No enterprise or commercial edition is offered.

No verified download or API-call volume is published by any primary source: neither the repository nor jina.ai/reader gives a request count or a user count, and the reader is consumed as a URL prefix rather than as a package, so there is no registry artifact to band. That leaves stars - 11,865 stargazers on the one declared repository, which bands at >10K stars, level 3, and a stars-based signal caps the level there because a star is not a use. The hosted r.jina.ai endpoint is widely used but its traffic is not published anywhere. The page the star count comes from carries a live counter, so its content will differ on any later fetch; that is expected drift on this source rather than a signal about the claim.

• https://github.com/jina-ai/reader recorded 2026-08-13

  Repo page, 11,865 stargazers, Apache-2.0, "Convert any URL to an LLM-friendly input with a simple prefix https://r.jina.ai/". No download, request or user figure.

A solid single-purpose reader and scraper, covering several document formats and adding a search mode. That puts it mid-tier on the dimensions that matter here - coverage, freshness, structured output and rate limits - against full agentic browse and search platforms.

• https://github.com/jina-ai/reader recorded 2026-08-13

  Feature list - HTML, PDF and MS-Office input, image captioning, markdown output, search mode via s.jina.ai, stateless operation with optional bucket caching.