Keyword search is how most people find papers, and it’s also how they miss them. Type the wrong term and a foundational study never surfaces. ResearchRabbit (researchrabbitapp.com) takes a different route: instead of matching words, it follows citations. Add a paper you already rate as a “seed”, and it builds an interactive visual map of the work around it — papers that cite your seed, papers your seed cites, and semantically similar research — then keeps recommending more as your collection grows.

The pitch that stuck is “Spotify for papers”: build playlists of research, follow the threads, and let the algorithm surface things you’d never have searched for. It’s earned hundreds of thousands of users among PhD students and academics, and it’s genuinely addictive in the best way — one paper leads to five more, and a field’s structure starts to reveal itself in front of you.

How ResearchRabbit works

Visual citation mapping

This is the heart of it. ResearchRabbit analyses citation relationships and renders them as a network graph, so you can see how papers connect rather than scrolling a list. Clusters reveal sub-fields; well-connected nodes reveal foundational work; sparse edges can reveal gaps. For understanding the shape of a research area quickly, the visual approach beats a results page hands down.

Collections and recommendations

You organise papers into collections — the “playlists” — by topic or project. As a collection grows, ResearchRabbit’s recommendation engine learns what you’re interested in and surfaces increasingly relevant suggestions. The 2026 interface adds search-path history, so you can retrace exactly how you stumbled onto a paper during a long exploration session and never lose the thread.

Authors, alerts and collaboration

Click an author to see and visualise their work, set up email digests that alert you when new papers relevant to your collections appear, and share collections with collaborators in real time. It’s a discovery and light project-management tool, not just a search box.

Fits your existing workflow

ResearchRabbit integrates with Zotero (one-way import) and supports BibTeX and RIS export, so it slots in alongside the reference manager you already use rather than asking you to switch. The team is explicit about a spam-free, quality-first approach and relies on public academic databases.

How ResearchRabbit scores

Overall score: 8.0 / 10, the average of the eight categories above.

ResearchRabbit pricing

ResearchRabbit was famously 100% free for years. That changed in late 2025: following its acquisition by Litmaps, the tool moved to a freemium model with a premium RR+ tier, while keeping its founding promise that the core stays free.

Note: some older reviews still describe ResearchRabbit as entirely free with no premium tier — that information predates the late-2025 RR+ launch. The core remains free to use.

Where it falls short

No reader, no summaries

ResearchRabbit is purely about discovery and mapping. There’s no built-in PDF reader or annotation, and no TLDR-style summaries like Semantic Scholar provides — once you’ve found a paper, you still head to Zotero, Mendeley or the publisher to actually read it. It finds the literature; it doesn’t help you digest it.

Echo chambers and coverage gaps

Because recommendations lean on citation networks, the algorithm can drift toward an echo chamber, repeatedly surfacing popular papers while quieter but relevant work stays hidden. Coverage is strong for STEM but thinner for humanities and social sciences, and the underlying databases can lag for very recent publications. The interface, for all its appeal, can also feel overwhelming to a first-time user.

Pros and cons

Who should use ResearchRabbit?

ResearchRabbit is ideal for PhD students and academics running literature reviews, anyone trying to map an unfamiliar field quickly, and researchers hunting for gaps where a topic is heavily referenced but lightly studied. If you think visually and want discovery to feel like exploration rather than a chore, it’s a joy to use — and the core costs nothing.

Pair it with the right complement. Use it for discovery, then Semantic Scholar for broad search with TLDR summaries, Scite to check whether key findings hold up, and Consensus or Elicit when you need synthesis or extraction. Visual mapping rivals like Connected Papers and Litmaps (its new owner) cover similar ground if you want to compare.

Verdict

ResearchRabbit nails one job and nails it beautifully: visual, citation-led discovery that surfaces the papers keyword search leaves behind. The collection workflow is clever, the recommendations genuinely useful, and the experience addictive in a way research tools rarely are — all with a core that remains free.

It sits just below the broader Semantic Scholar in our ranking because it’s deliberately narrow: no reader, no summaries, occasional echo chambers, and coverage that thins outside STEM. As one piece of a workflow rather than the whole thing, though, it’s outstanding — and for mapping a field fast, few tools are more enjoyable to use. Score: 8.0/10.

Frequently asked questions