Open Library API: Search for books and authors

The Open Library API provides access to one of the world's largest open book databases. It is free to use and requires no API key or account.

This guide shows you how to complete the core research workflow:

Search for a book by title, author, or keyword.
Retrieve full metadata for a specific work.
Look up an author's biography and other works.

Prerequisites

Before you begin, make sure you have the following:

An HTTP client. You can run the examples using curl, Postman, or any other client that sends HTTP requests.
An internet connection. No API key or account is required.
Familiarity with JSON. All API responses return JSON.

Base URLs

This guide uses two base URLs:

Service	Base URL
Search, works, authors, editions, subjects, ISBN	`https://openlibrary.org`
Cover images	`https://covers.openlibrary.org`

Add each endpoint path to the relevant base URL.

Any request can return 503 Service Unavailable if Open Library is temporarily down. Wait and retry, increasing the delay each time.

Step 1: Search for a book

Use the search endpoint to find a book by title, author, or keyword.

HTTP request

GET https://openlibrary.org/search.json

Parameters

Parameter	Example	Description
`q`	`q=republic`	Searches by general keyword.
`author`	`author=plato`	Filters by author name.
`title`	`title=republic`	Filters by title.
`limit`	`limit=5`	Sets the maximum number of results.

Text parameters (q, author, title) accept any value without validation, and unknown parameters are ignored rather than rejected. Numeric parameters like limit must receive a number, or the API returns 422 (see Error responses).

Send a request

The following request searches for the Republic:

curl "https://openlibrary.org/search.json?q=republic"

To filter, combine parameters — for example:

curl "https://openlibrary.org/search.json?author=plato&title=republic&limit=5"

Interpret the response

A successful request returns 200 OK with a JSON body. The following example shows a partial response for ?q=republic:

{
  "numFound": 165016,
  "start": 0,
  "q": "republic",
  "docs": [
    {
      "author_key": [
        "OL12823A"
      ],
      "author_name": [
        "Plato"
      ],
      "cover_i": 14418448,
      "ebook_access": "public",
      "edition_count": 794,
      "first_publish_year": 1554,
      "key": "/works/OL51831W",
      "title": "πολιτεία"
    }
  ]
}

Field	Type	Description
`numFound`	integer	Total number of matching works.
`docs`	array	Works matching your search query.
`docs[].author_key`	array	Unique identifiers for each author, as used internally by Open Library.
`docs[].author_name`	array	Author names, which often appear in non-Latin scripts.
`docs[].cover_i`	integer	A single cover image ID for this work. See Display cover images and author photos.
`docs[].ebook_access`	string	Public accessibility status of the full text. See the following table for possible values.
`docs[].edition_count`	integer	Number of published editions. A higher count generally reflects a longer publication history.
`docs[].key`	string	Unique identifier for the work. You use this in Step 2.
`docs[].title`	string	Title of the work, which often appears in a non-Latin script.

Handle non-Latin characters. Open Library stores text fields, including author_name and title, exactly as they appear in its database, often in their original scripts. For example, searching for "republic" may return "title": "πολιτεία" — the ancient Greek form. Make sure your application can handle and display non-Latin characters in these fields.

Understand ebook_access values.

| Value | Description | |-----------------|--------------------------------------------------------- -| | public | Fully accessible with no restrictions. | | borrowable | Available through the Internet Archive lending program. | | printdisabled | Accessible only to users with print disabilities. | | no_ebook | No digital version available |

Ancient Greek and Latin works are in the public domain, but modern translations are not. A recent English translation of the Republic likely returns "ebook_access": "borrowable", even though Plato's original text is thousands of years old.

Save the key value from your first result. It looks like /works/OL51831W — keep just the ID at the end (OL51831W). You'll use it in the next step.

Error responses

Status code	Cause	What to do
`422 Unprocessable Entity`	A parameter value has the wrong type (for example, a non-numeric `limit`).	Pass a number to numeric parameters like `limit`.

Step 2: Retrieve a specific work

Use the work key from Step 1 to retrieve its full metadata.

HTTP request

GET https://openlibrary.org/works/{work_id}.json

Replace {work_id} with the ID you saved — for example, OL51831W.

Send a request

curl "https://openlibrary.org/works/OL51831W.json"

Interpret the response

A successful request returns 200 OK. The following example shows a partial response for the Republic (OL51831W):

{
  "description": "The Republic is Plato's most famous work and one of the seminal texts of Western philosophy...",
  "authors": [
    {
      "author": { "key": "/authors/OL12823A" },
      "type": { "key": "/type/author_role" }
    }
  ],
  "covers": [14418448],
  "first_publish_date": "1963",
  "key": "/works/OL51831W",
  "subject_people": ["Plato", "Platōn"],
  "subject_times": ["Early works to 1800", "Ouvrages avant 1800"],
  "subjects": ["Political science", "Justice"],
  "title": "πολιτεία"
}

Field	Description
`description`	Summary of the work. For the format, see the note below.
`authors`	Author references. Each contains a `key` you use in Step 3.
`covers`	Cover image IDs. See Display cover images and author photos.
`subject_people`	Historical figures referenced in the work.
`subject_times`	Historical periods associated with the work.

Handle two description formats. The description field returns either a plain string or an object:

"description": "The Republic is Plato's most famous work..."

"description": {
  "type": "/type/text",
  "value": "The Republic is Plato's most famous work..."
}

Check the type before accessing the value; otherwise your code will fail on one of the two formats.

Handle duplicate subject_people and subject_times entries. Because both fields combine data from multiple library cataloging traditions, the same entity often appears more than once in different formats or languages:

"subject_people": [
  "Plato",
  "Platōn (427-347 p.Ch)",
  "Platon (0427?-0348? av. J.-C.)."
]

Before displaying these entries in your UI, de-duplicate or group them.

Save the author key from the work's authors array. Each entry's author.key looks like /authors/OL12823A. Keep just the ID (OL12823A) to use in the next step.

Error responses

Status code	Cause	What to do
`404 Not Found`	The work ID does not exist in Open Library.	Confirm the ID came from a Step 1 search response.

Step 3: Retrieve author details

Use the author key from Step 2 to retrieve the author's biography, dates, and external identifiers.

HTTP request

GET https://openlibrary.org/authors/{author_id}.json

Replace {author_id} with the ID you saved — for example, OL12823A.

Send a request

curl "https://openlibrary.org/authors/OL12823A.json"

Interpret the response

A successful request returns 200 OK. The following example shows a partial response:

{
  "alternate_names": [
    "Πλάτων",
    "Platon",
    "Platōn",
    "Platón",
    "Platone"
  ],
  "remote_ids": {
    "wikidata": "Q859",
    "project_gutenberg": "93"
  },
  "photos": [
    6457448,
    3366955,
    6319952,
    6319966,
    -1
  ],
  "death_date": "348/347 BC",
  "bio": "Classical Greek philosopher",
  "name": "Plato",
  "birth_date": "428/427 BC"
}

Field	Description
`alternate_names`	Other names the author is known by, including non-Latin scripts.
`remote_ids`	Cross-references to external databases. See the following table for details.
`photos`	Author photo IDs. See Display cover images and author photos. A value of -1 means no photo exists. Before building a URL, filter out this value.
`death_date`	Date of death. The format varies, and this field may be absent.
`bio`	Biography. This field returns either a plain string or an object, similar to the `description` field in Step 2.
`name`	Author's full name
`birth_date`	Date of birth. The format varies and may be approximate or absent for ancient authors.

Use remote_ids to link to external sources. This field is particularly useful if you're building a research tool — it lets you link an author profile to authoritative external databases without additional searching. For example:

Identifier	Description
`project_gutenberg`	Links to freely accessible texts on Project Gutenberg.
`wikidata`	Links to Wikidata's multilingual structured knowledge base.

Not every author has all identifiers populated. Before using a field, check if it exists.

Handle missing fields gracefully. For ancient authors, many fields — including bio, birth_date, and death_date — may be null or absent. Your application must handle both null and missing fields without throwing an error.

Retrieve all works by an author

Once you have the author key, retrieve all works associated with that author:

curl "https://openlibrary.org/authors/OL12823A/works.json"

This lets users start from a single text and discover an author's entire body of work in one request.

Error responses

Status code	Cause	What to do
`404 Not Found`	The author ID does not exist in Open Library.	Confirm the ID came from a Step 2 work response.

Next steps

Now that you've completed the core workflow, here are a few ways to build on it.

Explore subjects and topics

To browse all works in a specific category, use the subjects endpoint:

curl "https://openlibrary.org/subjects/ancient_greek_literature.json"

This enables category-level browsing, letting users explore an entire literary tradition without knowing specific titles.

Handle inconsistent subject taxonomy. Open Library's subject tags are community-contributed. The same body of literature may be cataloged under different tags depending on who added the record:

-greek_literature
-classical_literature
-ancient_philosophy
-ancient_greek_literature

For comprehensive results, run parallel requests across multiple subject endpoints and merge the results in your application.

Retrieve editions of a work

A work is the abstract text (for example, the Republic). An edition is a specific published version, such as an Oxford Classics translation with a particular ISBN and page count.

To retrieve all editions of a work, use the work key from Step 2:

curl "https://openlibrary.org/works/OL51831W/editions.json"

Use the work key, not the author key. Passing an author key to this endpoint returns 404 Not Found error. Before constructing a request, verify which key type you are using.

Display cover images and author photos

Both works and authors can have associated images. Works have cover images (from the covers field in Step 2); authors have photos (from the photos field in Step 3). Both use the same image endpoint and ID pattern.

GET https://covers.openlibrary.org/b/id/{id}-{size}.jpg

Replace {id} with a numeric ID from the covers array (Step 2) or the photos array (Step 3). Replace {size} with one of the following values:

Value	Description
`S`	Small
`M`	Medium
`L`	Large

Display a work's cover image. Use a numeric ID from the work's covers array (for example, 14418448):

curl "https://covers.openlibrary.org/b/id/14418448-M.jpg" --output cover.jpg

Display an author's photo. Use a numeric ID from the author's photos array (for example, 6457448):

curl "https://covers.openlibrary.org/b/id/6457448-M.jpg" --output photo.jpg

Handle missing photos. Unlike covers, the photos array can contain -1 as a placeholder when no photo exists:

"photos": [6457448, 3366955, -1]

Before constructing an image URL, filter out any -1 value. Otherwise, the request returns a generic "image not found" placeholder instead of failing cleanly.

Search by ISBN

If a user already knows a book's ISBN, you can retrieve the specific edition directly:

curl "https://openlibrary.org/isbn/9780140275360.json"

This returns the edition associated with that ISBN, which is more precise than a keyword search when the exact publication matters.

Handle missing and inconsistently formatted data

As you saw in Step 2 and Step 3, some fields are inconsistently formatted or absent entirely. Before deploying your application, make sure it handles the following cases:

Case	Fields affected
A field returns a plain string or an object.	`description`, `bio`
A field is `null` or absent entirely.	`bio`, `birth_date`, `death_date`, `covers`, `description`
A work has no associated authors.	`authors`
Fields contain duplicates in multiple languages.	`subject_people`, `subject_times`

Go further

For the complete endpoint reference, see the Open Library API documentation.

Prerequisites​

Base URLs​

Step 1: Search for a book​

HTTP request​

Parameters​

Send a request​

Interpret the response​

Error responses​

Step 2: Retrieve a specific work​

HTTP request​

Send a request​

Interpret the response​

Error responses​

Step 3: Retrieve author details​

HTTP request​

Send a request​

Interpret the response​

Retrieve all works by an author​

Error responses​

Next steps​

Explore subjects and topics​

Retrieve editions of a work​

Display cover images and author photos​

Search by ISBN​

Handle missing and inconsistently formatted data​

Go further​

Prerequisites

Base URLs

Step 1: Search for a book

HTTP request

Parameters

Send a request

Interpret the response

Error responses

Step 2: Retrieve a specific work

HTTP request

Send a request

Interpret the response

Error responses

Step 3: Retrieve author details

HTTP request

Send a request

Interpret the response

Retrieve all works by an author

Error responses

Next steps

Explore subjects and topics

Retrieve editions of a work

Display cover images and author photos

Search by ISBN

Handle missing and inconsistently formatted data

Go further