Is Heptabase a second brain app or a whiteboard tool?

Both. Heptabase is a note tool whose primary organising surface is an infinite canvas rather than a folder tree — cards you place and group spatially instead of filing into hierarchies.

Do both capture YouTube videos?

Yes, both transcribe YouTube. The difference is what happens next: Heptabase gives you a card to place on a canvas; BrainTube indexes it for semantic retrieval and exposes it over MCP so any AI client can query it.

Heptabase Pro at $8.99/mo (yearly billing) undercuts BrainTube Pro at $19, and sits near BrainTube Starter at $9. But BrainTube has a $0 tier and Heptabase's pricing page lists none, so at zero spend the comparison is not close. Above that they are priced similarly for different things — Heptabase meters AI usage, BrainTube meters capture.

Can I use both together?

Yes, and it is a sensible setup. Use Heptabase for the current project you are actively thinking through on a canvas, and BrainTube as the queryable archive underneath — everything you have consumed, retrievable from Claude, ChatGPT and Cursor via MCP.

Which is better for academic research?

Honest answer: Heptabase for actively working through a defined set of papers where seeing the spatial structure of your argument matters; BrainTube for a multi-year corpus you need to query and cite from across tools. They solve different halves of the research problem.

Can I move my data out of either?

BrainTube ships full JSON export on every plan including Free — source URLs, transcripts, highlights, notes. For Heptabase, export options should be checked against their current documentation rather than asserted here.

Comparison

BrainTube vs Heptabase

Heptabase is where you think through material by arranging it spatially. BrainTube is where what you consumed becomes queryable from any AI. Both handle YouTube transcripts and PDFs; the fork is who does the organising, and whether an outside model can read the result.

The distinction is authored structure vs derived structure. Heptabase's canvas encodes YOUR judgement about how ideas relate — a spatial argument you built by hand, which similarity scoring cannot reproduce. BrainTube's graph is derived from embeddings and entity extraction at ingest, which means it scales to material you never processed. Neither is better; they answer different questions. The question to ask is whether your bottleneck is thinking or retrieval.

Quick verdict

Pick Heptabase if your work is research-shaped — literature reviews, complex arguments, design thinking — and seeing the spatial structure of your ideas is the point. Pick BrainTube if your intake volume exceeds what you can process by hand and you need it queryable from Claude, ChatGPT and Cursor rather than inside one app.

Capability	BrainTube	Heptabase
Captures YouTube transcripts	Yes	Yes
PDF upload, highlights & annotation	Yes	Yes
Visual canvas — spatial arrangement of ideas	No	Yes
Capture without manual filing or arranging	Yes	No
Structure formed automatically from content	Yes	No
In-app AI chat over your material	Yes	Yes
AI usage metered by credits	Capture credits	AI credits — 100 to 8,100/mo by tier
Native MCP server — query from Claude, ChatGPT, Cursor	Yes	No
Spaced-repetition resurfacing (FSRS-5)	Yes	No
Real-time collaboration	No	Yes
Free plan	Free · 30 credits/mo	No free tier listed
Paid entry price	Starter $9 · Pro $19 · Pro+ $49	Pro $8.99 · Premium $17.99 · Premium+ $53.99

yes · partial · no

Where Heptabase wins honestly

Spatial thinking is a real cognitive tool. Best-in-class for literature reviews and complex arguments where seeing the shape of your ideas matters as much as the ideas themselves.
Authored structure encodes judgement. How you place and group cards is an argument in itself — one that similarity scoring cannot reproduce, because it is not in the content.
Genuinely strong PDF workflow. Unlimited uploads and OCR parsing on Premium, built for people who live in papers.
Real-time collaboration. Unlimited collaborator invites even on the entry tier.

Where BrainTube wins

MCP-native, portable across every AI client. Your library is queryable from Claude, ChatGPT, Cursor and any MCP-compatible tool — not just from inside one app.
Capture scales past what you can hand-arrange. A canvas with four hundred cards stops being a canvas. BrainTube's derived structure keeps up when volume outruns your patience for filing.
FSRS-5 spaced repetition. Material comes back on a schedule instead of sitting in an archive you forgot you had.
Free tier at $0, plus full JSON export on every plan. No paywall on either trying it or leaving.

Arranging vs retrieving

These are not really competitors so much as different stages of the same workflow. Heptabase is for the material you are actively thinking about — the ten papers on your desk this month, the argument you are building. BrainTube is for everything you consumed and might need later — the two thousand videos, podcasts and articles that make up your working context. Running both is a coherent setup: Heptabase for the current project, BrainTube as the queryable archive underneath it.

Does Heptabase support MCP?

As of July 2026 there is no publicly documented MCP endpoint for Heptabase. If that changes, this page will be updated. Until then, exposing your library to Claude, ChatGPT and Cursor over the Model Context Protocol is a BrainTube-only feature between the two.

Frequently asked

Is Heptabase a second brain app or a whiteboard tool?: Both. Heptabase is a note tool whose primary organising surface is an infinite canvas rather than a folder tree — cards you place and group spatially instead of filing into hierarchies.
Do both capture YouTube videos?: Yes, both transcribe YouTube. The difference is what happens next: Heptabase gives you a card to place on a canvas; BrainTube indexes it for semantic retrieval and exposes it over MCP so any AI client can query it.
Which is cheaper?: Heptabase Pro at $8.99/mo (yearly billing) undercuts BrainTube Pro at $19, and sits near BrainTube Starter at $9. But BrainTube has a $0 tier and Heptabase's pricing page lists none, so at zero spend the comparison is not close. Above that they are priced similarly for different things — Heptabase meters AI usage, BrainTube meters capture.
Can I use both together?: Yes, and it is a sensible setup. Use Heptabase for the current project you are actively thinking through on a canvas, and BrainTube as the queryable archive underneath — everything you have consumed, retrievable from Claude, ChatGPT and Cursor via MCP.
Which is better for academic research?: Honest answer: Heptabase for actively working through a defined set of papers where seeing the spatial structure of your argument matters; BrainTube for a multi-year corpus you need to query and cite from across tools. They solve different halves of the research problem.
Can I move my data out of either?: BrainTube ships full JSON export on every plan including Free — source URLs, transcripts, highlights, notes. For Heptabase, export options should be checked against their current documentation rather than asserted here.

Save once. Query anywhere.

For every AI you'll use.

Start free