Watch: The object-first model has a learning curve; users who...
Capacities
Capacities is object-based PKM where every note is a typed object with...
Monthly Free Annual Pro/Believer pricing is region-rendered
Best plan
Free
Risk: The object-first model has a learning curve; users who...
Editorial · no paid placements
Should you use it?
Capacities is object-based PKM where every note is a typed object with properties. The June 29 check is product scope, not only pricing: Pro now matters more for AI Chat Connectors 2.0, provider choice, richer AI context, image analysis, recurring tasks, and media-aware workflows, while Bulk Import is available to all users. Pick it for structural thinking and clean auto-links. Skip it for teams, plugin depth, or plain-Markdown portability.
- Buy if Solo researchers and writers who think structurally
- Pick Free; Pro/Believer pricing is region-rendered
- Skip if Teams needing real-time collaboration
Plan guidance
What to buy
Region-rendered
The object-first model has a learning curve; users who...
Current pricing source: Capacities pricing
Fit
Use it for this, skip it for that
Best for
- Solo researchers and writers who think structurally
- Knowledge workers linking sources with metadata
- Project and character trackers across connected objects
- Users wanting structure without Obsidian's setup tax
Avoid if
- Teams needing real-time collaboration
- Users requiring plain-Markdown portability
- Workflows dependent on heavy plugin ecosystems
- Watch out
- The object-first model has a learning curve; users who mainly want plain Markdown files or maximum plugin extensibility may prefer Obsidian.
Recent changes
Only what affects the decision
- Free / Pro / Believer
Rechecked for the Capacities vs Notion AI...
Capacities pricing - Free / Pro / Believer
Pricing remains region-rendered in public...
Capacities pricing - Free / Pro / Believer
Rechecked pricing, AI assistant docs, AI Chat Connectors docs, Release 65, and the What's New...
Capacities pricing
Alternatives
Best swaps
AI meeting assistant with unlimited free recording and transcription. Premium $20/mo, Team $19/user/mo, Business $34/user/mo add
$0-$34/user/month · 8.5/10 ObsidianLocal-first markdown knowledge base with 5,085 community plugins, including 580 AI-category plugins. AI arrives via Smart Connec
$0-$8/month (add-ons) + $50/year commercial · 8.5/10 Readwise ReaderAI-powered read-later app that ingests articles, newsletters, PDFs, ebooks, YouTube, and Twitter into one inbox with Ghostreader
$9.99/month annual or $12.99/month monthly for Reader + Readwise Full · 8.5/10Capacities comparisons
See all →Proof and score math Verified Jun 29
Proof
Why this recommendation is trusted
- Source
- Registered source
- Freshness
- Review due
- Confidence
- Low confidence
- Verified
- Review
- Volatility
- Volatile
Stale source capacities-release-66.
Editorial score
Unweighted average of 4 axes · confidence high
- Utility 7/10
How much real work it can do for a competent operator, end to end.
- Value 8/10
What you get for the dollar relative to the closest alternative.
- Moat 7/10
How hard it would be for a competitor to replicate the underlying advantage.
- Longevity 7/10
How likely the product is to still be best-in-class 24 months out.
Verified facts
- Best For Capacities is best for personal knowledge management built around typed objects, properties, backlinks, and graph-like organization rather than folders or freeform note piles.
- Pricing Anchor Capacities pricing should be evaluated against Pro AI access, image analysis, unlimited media uploads, sync needs, regional currency, and whether the object-based model will become a daily workflow. The current text crawl confirms the core product is free, but the detailed Pro and Believer plan grid remains region-rendered and should be verified in checkout.
- Watch Out For The object-first model has a learning curve; users who mainly want plain Markdown files or maximum plugin extensibility may prefer Obsidian.
- Data Model The defining product choice is object-based notes: people, books, projects, meetings, and other entity types can carry properties and reusable views.
- Ai Surface Capacities AI now includes richer note context, saved AI chat objects, AI commands, AI auto-tagging, AI property auto-fill, web search for Capacities AI, Pro image analysis, BYOK-compatible note reading/searching, MCP AI Chat Connectors, model-provider selection, and Release 66 connector write actions.
- Connector Surface AI Chat Connectors are a Pro AI-enabled surface that lets ChatGPT, Claude, Cursor, Le Chat, and other MCP-compatible tools search notes, read objects, save text, link objects, and, as of Release 66, create pages/tasks/custom objects, update object properties, and append content to objects.
- Recent Product Change Release 66 added AI Chat Connectors 2.0 write capabilities, concurrent AI chats, AI media-section reading improvements, grouped source cards, mobile linking fixes, editor polish, import fixes, and performance cleanup. Release 65 remains the prior bulk-import and Pro image-analysis milestone.
Full review notes Long-form details, FAQ, and source history
An object-based personal knowledge management app built by Capacities GmbH. Every note is a typed object (Book, Person, Project, Daily Note, or custom type) with defined properties that form a connected knowledge graph. The Pro tier now matters for AI assistant features, MCP AI Chat Connectors, model-provider choice, richer context, image analysis, and media-aware knowledge work.
The free-core promise is still visible on the pricing page, but detailed plan pricing and some limits render by region and account context. Treat older fixed USD/NZ/EUR examples as budgeting hints only. Verify Pro and Believer pricing, AI budget, media limits, and billing currency inside your live checkout before buying.
What Changed Since The Last Refresh
- Release 64 added recurring tasks with schedule/completion-date modes, catch-up behavior, excuse/skip options, task stats, and a recurring task dashboard.
- Release 64 also added AI model-provider choice across OpenAI, Gemini, Anthropic, Mistral, and xAI, with model-picker details for EU-hosted models and usage cost.
- Release 65 made Bulk Import available to all users, brought image analysis to Capacities Pro, improved recurring task edge cases, reworked the side panel, and added AI web search and AI chat polish.
- The What’s New index now lists May 2026 v1.65.x continuous improvements above Release 65, including small fixes, query suggestions for object dashboards, tags view for task dashboards, and an AI panel fix.
- Release 66 added AI Chat Connectors 2.0, allowing connected assistants to create pages and tasks, create custom objects, update object properties, and append content to objects. It also improved concurrent AI chats, media-section reading, source-card display, mobile linking, imports, and performance.
- Release 62 and the current docs keep AI Chat Connectors as a Pro beta using MCP, so ChatGPT, Claude, Cursor, Le Chat, and other compatible tools can search notes, read objects, add daily-note content, and link Capacities objects.
- Release 63 added Related Content, deadlines, Search 3.0, AI chat upgrades, and full Bulk Import access for Believers.
- Release 61 added AI media analysis for images, including searchable descriptions, OCR-like extraction, and media-specific object categories.
- Pricing did not become easier to quote: the public text crawl confirms the free-core promise, but exact Pro and Believer plan prices should still be verified in checkout because the plan grid is region-rendered.
Recent Changes
- May 2026 continuous improvements: The What’s New index now shows v1.65.13 and v1.65.6 above Release 65, including small fixes, object-dashboard query suggestions, task-dashboard tags view, standardized task context picker behavior, and an AI panel fix.
- Release 65 (May 2026): Bulk Import is available to all users, Image Analysis is available for Pro users, recurring tasks gained more scheduling polish, the side panel was reworked, and AI web search was added to the assistant.
- Release 64 (May 2026): Recurring tasks shipped, and Capacities AI added model-provider choice across OpenAI, Gemini, Anthropic, Mistral, and xAI.
- Release 63: Related Content, deadlines, Search 3.0, AI chat upgrades, and full Bulk Import access for Believers.
- Release 62: AI Chat Connectors arrived as a Pro beta, integrating Capacities with ChatGPT, Claude, Cursor, Le Chat, and other MCP-compatible clients.
- Release 61: Image and media analysis with new object categories, OCR-style extraction, and color preview.
- Release 59: AI Assistant gained richer context across notes and showed which notes it used to answer questions.
- Release 58: Search 2.0 rebuilt fully on-device for speed and privacy.
System Verdict
Pick Capacities if structure is the point. Object typing forces intentionality at the moment of capture. Every Book gets an author property, every Project gets a status, every Person gets links to meetings. That structural prior is what Obsidian and Notion leave to the user.
Skip it for team work or plain-Markdown portability. No real-time collaboration, no shared workspaces. Export exists but the proprietary database structure loses fidelity on the way out. If you might leave, Obsidian is a safer long-term home.
Who pays which tier: Free for indefinite solo use without AI, Pro for the AI assistant, AI Chat Connectors, provider choice, and richer media/AI workflows, Believer for users who want to support development and get early features such as Bulk Import access. Verify your live regional checkout before budgeting from an old fixed-price quote.
Key Facts
| Core model | Typed objects with custom properties |
| Pre-built object types | Book, Person, Project, Daily Note, Article, Video, Podcast |
| Custom objects | Listed as unlimited on the current Free tier |
| AI assistant | Writing, chat over notes, context selection, web search, image analysis, auto-fill properties, auto-tagging, AI chat objects |
| AI backend | Provider choice across OpenAI, Gemini, Anthropic, Mistral, and xAI, plus BYOK paths |
| External AI connectors | Pro beta via MCP for ChatGPT, Claude, Cursor, Le Chat, and compatible tools |
| Graph view | Visual map of object connections |
| Platforms | Web, macOS, Windows, iOS, Android |
| Offline | Full, with sync on reconnect |
| Encryption | End-to-end on sync |
| Free tier | Core product remains free; verify media and plan limits in live regional pricing |
| API | None public |
Current AI connector, pricing, and product-change data points were verified against vendor docs on 2026-06-29. See Sources.
What it actually is
A typed-object database dressed as a notes app. Creating a note starts with picking an object type. Each type carries properties that auto-link across the graph: a new Book links to its Author (a Person object), the Person links back to every Book and Project they appear in.
The AI assistant layers on top. Writing aids generate or rewrite text. AI Chat can search notes, read relevant content, use backlinks, create saved AI chat objects, and work with selected context. Auto-fill can populate object properties. AI Chat Connectors let external tools reach into a selected Capacities space through MCP, which is powerful but also makes consent, space selection, and data-sharing review more important.
The moat: no other mainstream PKM tool enforces object typing at the capture step. The ceiling: the proprietary format limits export fidelity, and there is no plugin marketplace to extend behavior.
When to pick Capacities
- You think in systems. Researchers tracking sources, authors tracking characters, founders tracking decisions all benefit from typed objects.
- You want structure without Obsidian’s setup cost. Capacities ships the object system on day one. No plugin stack to assemble.
- Daily notes plus object links is the habit. Daily notes auto-link mentioned objects so a journal entry becomes part of the graph.
- Free tier covers the need. The free-core promise is still the safest starting point for solo users.
- You want AI tools to see your PKM. Pro’s AI Chat Connectors make Capacities more useful beside ChatGPT, Claude, Cursor, Le Chat, and other MCP-compatible tools.
- E2EE sync matters. Cross-device sync is end-to-end encrypted.
When to pick something else
- Local-first with plugins: Obsidian. Free, portable Markdown, 1,000+ plugins.
- Team wiki with databases: Notion AI. Real-time collab, full database depth.
- AI-first capture without typing: Mem. Skip the object step entirely.
- Cited document Q&A: NotebookLM. Free, citation-grounded research.
- Reading-plus-retention: Readwise Reader. Ingestion and spaced-repetition review.
Pricing
| Plan | Price | Key Limits |
|---|---|---|
| Free | $0 core product | Best for testing the object model before paying. Verify current media limits on the pricing page. |
| Pro | Region-rendered | AI assistant, AI Chat Connectors beta, provider choice, richer AI workflows, and Pro features. |
| Believer | Region-rendered | Everything in Pro plus support for development and early/beta access such as Bulk Import. |
Prices verified 2026-06-29 via capacities.io/pricing. The text crawl confirms the free-core promise, but detailed Pro and Believer prices are region-rendered, so verify your account region before budgeting.
Against the alternatives
| Capacities Pro | Obsidian | Notion AI | Reflect | |
|---|---|---|---|---|
| Price | Region-rendered Pro | Free | $10-$20/mo | $10/mo |
| Storage format | Typed objects | Local Markdown | Pages + databases | Markdown + backlinks |
| AI built-in | Yes, Pro+, with MCP connectors and provider choice | Via plugins | Yes | Yes |
| Offline | Full | Full | Limited | Full |
| E2EE sync | Yes | Local only | No | Yes |
| Team collaboration | None | None | Real-time | None |
| Plugin ecosystem | None | 1,000+ | Limited | None |
| Best viewed as | Structured PKM | Power-user vault | Team wiki | Privacy-first PKM |
Failure modes
- Proprietary storage. Export to Markdown/CSV exists but loses object relationships. Migration off Capacities is a one-way rebuild, not a file move.
- Learning curve on setup. Object types and properties need upfront design. Users who want to just write tend to bounce.
- AI still shares selected data with providers. Capacities docs say AI use can share data with the selected model provider. Review provider terms, BYOK settings, EU-hosted model options, image-analysis behavior, and connector consent before using sensitive notes.
- Connector access raises governance stakes. AI Chat Connectors are useful, but they let external AI tools search notes, read objects, add daily-note content, and link objects after consent.
- No real-time collaboration. Shared workspaces and team editing do not exist.
- Smaller ecosystem than Notion or Obsidian. Fewer templates, community resources, and integrations.
- No public API. Programmatic access and custom automations are limited.
- AI requires internet. Core notes work offline, but AI chat and property auto-fill need a live connection.
Methodology
This page was produced by the aipedia.wiki editorial pipeline, an automated system that ingests vendor documentation, verifies pricing and model details against primary sources, and generates the editorial analysis. No individual human wrote this review. Scoring follows the four-dimension rubric at /about/scoring/ (Utility x Value x Moat x Longevity, unweighted average). Last verified 2026-06-29 against capacities.io/pricing, Capacities AI assistant docs, AI Chat Connectors docs, Capacities Release 66, and the Capacities changelog.
FAQ
Is the free plan actually unlimited? The core product remains free according to Capacities’ pricing page. Verify current media limits, AI access, and regional plan details in live checkout before treating an old plan table as final.
Capacities vs Obsidian? Obsidian stores notes as local Markdown files with manual linking. Capacities uses typed objects in a proprietary database with property-driven auto-links and built-in AI. Obsidian suits tinkerers with strong file-ownership preferences. Capacities suits structured thinkers who want the object system out of the box.
Capacities vs Notion AI? Choose Capacities for solo object-based PKM. Choose Notion AI when the same notes need to become a shared team workspace with databases, meeting notes, search, permissions, and agents.
Does AI work offline? No. Writing aids and AI Chat require internet. Core note creation, editing, and browsing all work offline with sync on reconnect.
What’s the difference between Pro and Believer? Both include the AI assistant and Pro features. Believer is a support tier that funds ongoing development and gets early access to some features, such as Bulk Import. Pro is sufficient for most users unless early access matters.
Can Capacities handle teams? No. Capacities is solo-focused. No shared workspaces or real-time collaboration. Notion is the right tool for team knowledge bases.
Sources
- capacities.io/pricing: current Free, Pro, and Believer tiers
- Capacities AI assistant docs: AI assistant scope, provider data sharing, chat, commands, auto-tagging, and property auto-fill
- Capacities AI Chat Connectors docs: MCP connector scope and Pro beta status
- Capacities Release 65: Bulk Import for everyone, Image Analysis for Pro, recurring-task updates, side-panel changes, AI web search, and AI chat improvements
- Capacities Release 64: recurring tasks and AI model-provider choice
- Capacities changelog: recent feature releases
Related
- Category: AI Notes
- Comparisons: Capacities vs Notion AI · Capacities vs Obsidian
Reader reviews
Embed this score on your site Free. Links back.
<a href="https://aipedia.wiki/tools/capacities/" target="_blank" rel="noopener"><img src="https://aipedia.wiki/badges/capacities.svg" alt="Capacities on aipedia.wiki" width="260" height="72" /></a> [](https://aipedia.wiki/tools/capacities/) Badge value auto-updates if the editorial score changes. Attribution via the link is required.
Cite this page For journalists, researchers, and bloggers
According to aipedia.wiki Editorial at aipedia.wiki (https://aipedia.wiki/tools/capacities/) aipedia.wiki Editorial. (2026). Capacities: Editorial Review. aipedia.wiki. Retrieved July 2, 2026, from https://aipedia.wiki/tools/capacities/ aipedia.wiki Editorial. "Capacities: Editorial Review." aipedia.wiki, 2026, https://aipedia.wiki/tools/capacities/. Accessed July 2, 2026. aipedia.wiki Editorial. 2026. "Capacities: Editorial Review." aipedia.wiki. https://aipedia.wiki/tools/capacities/. @misc{capacities-editorial-review-2026,
author = {{aipedia.wiki Editorial}},
title = {Capacities: Editorial Review},
year = {2026},
publisher = {aipedia.wiki},
url = {https://aipedia.wiki/tools/capacities/},
note = {Accessed: 2026-07-02}
} Spotted an error or want to share your experience with Capacities?
Every tool page is re-verified on a recurring cycle, and corrections land faster when readers flag them directly. If you spot a stale fact, a missing capability, or have used Capacities and want to share what worked or didn't, the editorial desk reviews every message sent through this form.
Email editorial@aipedia.wiki