For literature reviews
A literature review canvas that shows the whole field at once.
Slashspace is a literature review canvas for papers, PDFs and web sources. In a chat scroll you see one thread and the last few messages. On a canvas every paper, question and draft sits side by side, so you can hold the whole field in your head, see where sources agree and disagree, and write the synthesis from all of it.
Praneeth Pike, founder of Slashspace (formerly RabbitHoles AI). Updated October 8, 2026.
$89/year, starting with a 3-day trial (card required). Or $249 one-time lifetime.
What is a literature review canvas?
A literature review canvas is an infinite workspace where each source, question and draft is a separate piece you can see and connect. Instead of asking one long chat about thirty papers, you give each research question its own chat, connect only the papers it should read, and pull the answers together in a final synthesis.
A literature review is an argument about a field, not a list of summaries. The hard part is holding many sources in mind at once and seeing how they relate. That is what a canvas is for.
From a folder of PDFs to a synthesis, in four steps
- Drop your papers on the canvas. Drag PDFs, Word files or slides onto the canvas. Each becomes a Document node. When it shows Ready, connected chats read its full converted text, not snippets.
- Give each research question its own chat. Start a chat per question, such as "How do studies measure X?" or "Where do the results disagree?". Connect only the papers that question needs, by dragging a connection or typing
@and picking the paper. - Branch when a source raises something new. Select a sentence in any answer and ask about it. A branch opens next to the chat it came from, with the passage quoted, so the side question never clutters the main thread.
- Write the synthesis from everything at once. Group the papers (or the finished question chats) and connect the group to a new chat. It reads everything inside the group in full, so you can draft the review from all of it without copy and paste.
Why a canvas beats a chat scroll for literature reviews
- See the whole field at once. Papers, questions and drafts sit side by side instead of scrolling out of view.
- Control what each answer is based on. A chat reads only what is connected to it. Click a chat and every source feeding it lights up, so you know exactly what an answer drew on.
- Keep questions apart. Sibling branches can't see each other, so the methods question never leaks into the findings question.
- Turn sprawl into a map. Tidy Up lays the canvas out in columns that follow your connections: sources on the left, the chats that use them in the middle, the synthesis on the right.
- Find anything later. Press
⌘K(Ctrl+Kon Windows and Linux) to search the canvas. AI Search can find text inside your documents.
Any model, per question
Pick a model for each chat. Use the hosted models, your own API keys, or local models through Ollama. If you already pay for Claude or ChatGPT, Slashspace drives the official Claude Code and Codex CLIs on your computer, signed in with your own account.
To check a claim, branch the same question to two or three models and compare the answers side by side on the same papers.
Your canvas stays yours
Every canvas is a folder on your computer. When you're done, download the canvas as Markdown and paste the conversations into your draft, or copy a single reply, table or passage.
Where Slashspace fits next to the tools you already use
A literature review usually needs more than one tool. Here is an honest split.
| Tool | Best at | Use it with Slashspace? |
|---|---|---|
| Zotero | Collecting, organizing, annotating and citing papers. Free and open source, with thousands of citation styles and apps for Mac, Windows, Linux, iOS and Android. | Yes. Keep your library and citations in Zotero, and bring the PDFs you're reading into Slashspace to think across them. |
| Gemini Notebook (formerly NotebookLM) | Answers grounded in the sources you upload, plus study outputs such as Audio Overviews, Video Overviews, Mind Maps and slide decks. | Yes. It's a strong choice when you want grounded answers from one set of sources. Slashspace is for when you want any model, a separate chat per question and all of it laid out on one canvas. |
| Elicit | Searching a large index of research papers, summarizing them and chatting with papers. It can import from Zotero and has a free Basic plan. | Yes. Use it to find and screen papers, then synthesize the shortlist on a canvas. |
| Slashspace | Holding many sources and questions in view at once, branching follow-ups, choosing the model per question and synthesizing across everything you connect. |
Slashspace is not a reference manager, a paper search engine or a citation generator. It's where the thinking happens between finding papers and writing the review. If you want paper search inside your chats, Agent mode chats can use research connectors such as Semantic Scholar, OpenAlex, Consensus and ChatAcademia.
Good to know before you start
- Documents are processed on Slashspace's servers. When you add a Document node, a copy is uploaded and converted to text there, so the file leaves your device. For sensitive material, paste the passages you need into a text note instead. Details: privacy
- Document nodes are marked beta in the right-click menu.
- Very long documents are skipped. If a converted document is larger than about 500 KB of text (very roughly 80,000 words), connected chats leave it out and say so. Split long theses or books into chapters.
- Every connected source is sent in full with every message. Connect big papers only to the chats that need them, or pick a model with a large context window.
- Desktop app for macOS, Windows and Linux.
Pricing
Every feature, unlimited canvases.
- Yearly: $89 a year, starting with a 3-day trial (card required).
- Lifetime: $249, one payment.
Frequently asked questions
Can AI write my literature review for me?
It can draft one, but a good review is your argument about the field. Slashspace is built for doing that thinking with AI: you decide which sources each question reads, compare answers across models and keep every step visible. Check every claim against the paper before it goes in your draft.
How many papers can I put on one canvas?
Canvases are unlimited on both plans. What matters is how much you connect to one chat: every connected paper is sent in full, so a synthesis chat wired to many long papers can go past a model's context window. Group papers by theme and synthesize each group first.
Does Slashspace cite sources?
Slashspace shows which papers each chat drew on through its connections, but it is not a citation manager. Ask the model to name the paper behind each point, then check it and cite it from your reference manager.
Is my research private?
Canvases live in a folder on your computer and aren't synced to Slashspace's servers. Documents, web pages and AI Search do send content to Slashspace's servers, and chats send their context to the model you pick (through Slashspace's servers when you use a hosted model). The privacy page lists exactly what goes where.
Can I use my own API key or a local model?
Yes. Add keys for providers such as OpenAI, Anthropic, Google or OpenRouter, or run local models through Ollama. Chats on your own keys and local models cost nothing extra on top of your plan.
Lay the whole field out and write from all of it.
Drop in your papers, give every question its own chat and see the review take shape on one canvas.