SciSpace API: Automate Research Paper Search & Literature Reviews

The SciSpace API brings research paper search and an AI research agent into your own tools. With one API call you can find the 100 most relevant papers on any question in seconds. You can also run a full literature review that comes back as a cited report, spreadsheet or slide deck.

The API is in early access, with public launch a few weeks away. This guide covers everything we shared in our preview webinar: the two APIs, when to use each, early pricing, and how to get your developer started.

Join the SciSpace API waitlist →

Early-access users get free credits and one-on-one onboarding at launch.

Why use a research paper API?

A research paper API removes the manual steps between asking a research question and using the answer. Today, even with AI research tools, the workflow looks like this:

1. Type a query and start a task.
2. Wait 20 to 30 minutes for a long task to finish, then come back to check.
3. Review the results, download them, and save them to Google Drive or your reference library.
4. Repeat for every question, every week.

There is also a ceiling on scale. A person can only run so many searches in parallel, and even SciSpace's highest plan caps you at 4 to 8 parallel tasks.

An API connects the steps. Your system sends a question, SciSpace finds and analyses the papers, and the output lands where you work: a Drive folder, a spreadsheet, a dashboard, or an AI assistant like ChatGPT or Claude. It can run on a schedule, so a daily search for new papers needs no manual work at all.

Search API: academic paper search in plain language

The Search API returns the top 100 ranked research papers for any natural-language question, in a few seconds. It does three things: finds papers that match your query, ranks them by relevance, and returns the top 100.

Ask "effects of climate change on biodiversity" and each result includes:

• Title and abstract
• Journal or conference name
• DOI
• Full-text PDF link, where the paper is open access

Key features

• Multilingual. Query in Japanese, Spanish or any other language and get relevant papers back.
• Filters as parameters. Limit results by publication year (for example, the last two years), open access only, or source type such as conference papers or patents.
• Fast and low cost. Search is the API's only job, so calls finish in seconds at the lowest price point.

The Search API queries the SciSpace database, which is updated monthly and built to be comprehensive.

What you can build with the Search API

• Natural-language search over your own library or paper database
• A scheduled job that finds new papers on a topic every day and emails them to your team
• Evidence for course pages: find the papers that back each claim in your content
• Search over internal documents, such as "What do our SOPs say about cleaning validation?"

Task API: an AI research agent you call by API

The Task API runs any research task you describe in plain language, from a single cited answer to a full systematic review. It is the same engine as the SciSpace Agent, available inside your own product.

Give it a task and it plans the work itself. A request like "Find recent papers on CRISPR off-target effects and write a report with citations" becomes a chain of sub-tasks:

1. Search for papers across databases
2. Re-rank the results
3. Read and analyse the papers
4. Write the report, with every claim backed by a cited paper

What the Task API can deliver

• Cited answers to research questions
• Literature reviews and systematic literature reviews
• Evidence tables comparing many papers in a spreadsheet
• Data analysis on a dataset you upload
• Slide decks, posters, and image outputs
• Peer-review reports and template-based reports

Open-ended or predefined. Run any task your way, or use a predefined pipeline. The systematic literature review pipeline follows a fixed protocol of steps instead of working them out at run time, so results stay consistent.

Any output format, with follow-ups. Ask for a Word file, Markdown, CSV or presentation and the response includes a link to the file. You can also follow up on a result, for example asking it to shorten a 10-page report to 2 pages.

Unlike the Search API, the Task API searches beyond the SciSpace database, pulling papers from PubMed, arXiv, Google Scholar and more.

Search API vs Task API: which should you use?

Use the Search API when you only need a ranked list of papers; use the Task API when you need an answer, review or report produced for you.

Choose the Search API if you need papers or passages rather than an answer, results must appear instantly, your own system reads and screens the papers, or you run high volumes at the lowest cost.

Choose the Task API if you want a summary, cited answer, review or report generated for you, or need the widest possible coverage across databases.

Or combine them. Many teams will use both: the Search API to shortlist papers cheaply, then the Task API to analyse the shortlist. That balances coverage, quality and cost.

Workspaces, pricing and developer tools

Free workspaces for your papers

No database of your own? Upload 10,000 or 20,000 papers to a SciSpace workspace and run the Search or Task API on top of them. Workspaces work like Google Drive, with folders and files, and storage is free with any API usage. You can also connect your own database instead.

Pay-as-you-go API pricing

There is no subscription or plan to choose. You pay only for what you run, from one balance shared by search and tasks.

Every new account gets $5 in free credits, and we can top up credits for testing on request. Auto-recharge keeps production systems running without balance checks. Prices are early estimates and may change before launch.

Easy to set up

Your developer gets clear API docs covering every input, output and parameter. There is a no-code playground to try queries and inspect responses, and a ready-made prompt to paste into Claude Code or Codex to start testing right away. A dashboard shows usage and cost per call.

Reserve early access and free credits →

What you can build with the SciSpace API

The SciSpace API fits anywhere research questions turn into documents, alerts or decisions.

Not technical? How to brief your developer

You don't need to write code to use the SciSpace API. Give your developer answers to these five questions and they can do the rest:

1. Outcome: What do you want? For example, a weekly table of new papers on a topic.
2. API: Search, Task, or both? Describe the use case and your developer can recommend the mix.
3. Sources: Published papers, web results, files in your workspace, or your own database?
4. Format: Report, spreadsheet, CSV or presentation?
5. Frequency and budget: How often should it run, and what can you spend each month?

How to get started with the SciSpace API

1. Join the waitlist and tell us your use case.
2. Get access at launch, with $5 in free credits and one-on-one help from our team.
3. Try the playground, brief your developer, and start building.

Join the SciSpace API waitlist →

We'll reach out personally as soon as the API launches.

Frequently asked questions

Watch the full webinar

The 31-minute preview walks through both APIs, the pricing model and use-case examples. For more, see the SciSpace AI literature review tool and Chat with PDF.

注意转载授权 · 阅读原文