Skip to main content

◆ EPISODIC_MEMORY_LAYER

What happened, where, and when. Always.

Years of scattered notes → one structured memory. Ravenbase captures, structures, and surfaces everything you've ever known.

◆ NO_CREDIT_CARD_REQUIRED

◆ THE_PROTOCOL

Three steps to permanent memory

01◆ UPLOAD

Capture everything

Drop in notes, chat exports, PDFs, or paste raw text. Ravenbase accepts any format without judgment.

02◆ STRUCTURE

Automatic structuring

The AI extracts entities, dates, places, and relationships — building a typed knowledge graph from your raw material.

03◆ GENERATE

Surface and synthesize

Query your memory in natural language. Generate meta-documents. Get precisely cited answers across your entire history.

◆ SYSTEM_MODULES

Built for long-term thinkers

◆ MEMORY_INBOX

Memory Inbox

Conflicts become conversations, not silent overwrites. Review AI-detected contradictions and resolve them with a keystroke.

◆ META_DOCUMENTS

Meta-Documents

Years of context → one perfect document. AI-generated synthesis that cites every source from your knowledge graph.

◆ KNOWLEDGE_GRAPH

Knowledge Graph

Concepts connected across time, not just semantics. Navigate every entity, date, and relationship you've ever captured.

Feature Deep-Dive

◆ MEMORY_INBOX

React vs Vue conflict — 94% match
Job title update — 91%
Framework preference — 88%
ACCEPT
KEEP_OLD

◆ MEMORY_INBOX

Capture without friction, review with intent

The Memory Inbox separates capture from processing. Drop in anything — the AI structures it later. You stay in flow while Ravenbase quietly organizes everything you throw at it.

◆ META_DOCUMENT

SOURCES (3) →

notes_2022.md · chat_export.json · resume_2023.pdf

◆ META_DOCUMENTS

Synthesis documents that cite their sources

Every meta-document is generated from your actual knowledge graph. Every claim links back to a specific memory entry. Export to PDF for sharing, print for archiving.

◆ HALL_OF_RECORDS

Trusted by researchers, builders, and thinkers

REF-0042

“247 research papers, 3 years of reading notes — all queryable in under a second. I found a connection between two papers I'd completely forgotten about.”

Dr. Ananya Sharma

ML Research Lead, DeepMind

REF-0088

“Ravenbase caught a contradiction between our Q2 strategy doc and the Q4 investor deck. That kind of drift is invisible until someone gets burned by it.”

Marcus Lindgren

VP of Product, Series B Startup

REF-0117

“My thesis bibliography built itself from 4 years of reading notes. The meta-document feature synthesized 180 sources into a 3-page literature review draft in 40 seconds.”

Priya Okonkwo

PhD Candidate, Stanford NLP Lab

◆ FREQUENTLY_ASKED

Questions

◆ PRICING_TIERS

Memory that pays for itself

Start free. Upgrade when your knowledge graph outgrows the basics.

◆ FREE_TIER

Free

Free

500 credits/month

  • 10 sources source uploads
  • 10 / month Meta-Documents
  • 3 profiles
  • Claude Haiku only
Get started
◆ PRO_TIERMOST_POPULAR

Pro

$15/month

2,000 credits/month

  • Unlimited source uploads
  • Unlimited Meta-Documents
  • 20 profiles
  • Haiku + Sonnet
◆ TEAM_TIER

Team

$49/month

6,000 shared credits/month

  • Unlimited source uploads
  • Unlimited Meta-Documents
  • Unlimited
  • Haiku + Sonnet
  • Cross-user graph sharing (3 seats)
◆ FEATURE_COMPARISON
FeatureFREEPROTEAM
Source uploads10 sourcesUnlimitedUnlimited
Meta-Documents / month10 / monthUnlimitedUnlimited
Profiles3 profiles20 profilesUnlimited
AI model accessClaude Haiku onlyHaiku + SonnetHaiku + Sonnet
Memory Chat
Graph Explorer
Claude Sonnet access
Cross-user graph sharing (3 seats)

Your knowledge, structured forever.

Start building your permanent memory today. Free to try, no credit card required.

Start for free →