Cortex vs Reflect
Reflect is a beautiful, fast networked notebook with tasteful AI. Cortex shares the speed obsession but goes further on capture (voice-first), automation (agents), and depth (a self-organizing graph you can query in plain language).
Where Reflect shines
- Gorgeous, minimal design with genuinely fast performance
- End-to-end encrypted notes — a rare privacy stance
- Elegant daily-notes and backlinking flow
Where Cortex differs
- Capture: Reflect has voice transcription; Cortex goes a step further — transcription plus structure: title, tags, bullets, and extracted tasks, automatically.
- Agents: Reflect's AI assists inside a note. Cortex agents run across your whole graph on schedules — weekly reviews, meeting debriefs, topic watching.
- Retrieval: Cortex's Cmd+K chats with your entire knowledge base using semantic search, not just keyword recall.
Side by side
| Cortex | Reflect | |
|---|---|---|
| Capture | Voice-first — speak, get a structured note in <3s | Type or voice transcription (raw text) |
| Bi-directional links | Yes — [[wiki-links]] + auto-suggested connections | Yes — backlinks |
| Knowledge graph | Self-organizing, with vector semantic search | Basic graph |
| Autonomous agents | Yes — durable agents run on your notes 24/7 | No scheduled agents |
| Chat with your notes | Yes — chat with your whole graph (Cmd+K) | AI assistant in-note |
| Local-first / offline | Local-first with offline support | Cloud with offline cache |
| Privacy | E2E encrypted sync, never trains on your data | End-to-end encrypted |
| Bring your own key | Yes — bring your own OpenAI/Anthropic key | Partial (AI features) |
| Pricing | Free tier planned · Pro $15/mo | Paid subscription |
Who should stay with Reflect
Common questions
Both are E2E encrypted — what's the real difference?
Automation depth. Reflect assists while you write; Cortex structures your capture automatically and runs durable agents on your knowledge when you're not there.
Is Cortex as fast as Reflect?
That's the bar. Cortex is local-first with cached vaults — notes open instantly, offline included. Speed is a feature we refuse to trade away.
Can I migrate from Reflect?
Cortex is markdown-first, so anything you can export as markdown imports cleanly.
Try the Cortex wedge in 10 seconds
Speak a thought on the early-access page and watch it become a structured note.
Join the founding queue →Cortex is in early access. This page reflects our understanding of Reflect as of July 2026 — if we got something wrong, tell us and we'll fix it.