Get This Tool
Ultramemory
Pricing
- Model
- Free
Summary
The decision got made in Slack, the follow-up landed in email, the context lives in a Drive doc — and six weeks later you are reconstructing the thread from memory because search across three tools gives you fragments, not answers. Ultramemory indexes all of it into a single local database on your Mac and lets you ask plain-language questions with every answer pinned to the source it came from.
The tool pulls from Gmail, Slack, Google Drive, Calendar, local files, screenshots, and Apple Notes — OCR and transcription run on device, nothing touches a server. Retrieval works without any AI model installed: deterministic full-text and vector search return ranked results immediately. Add a local model through Ollama and answers become conversational, with the model reading only the evidence it cites. The citation trail is the differentiating mechanic — click any claim and the original message or file opens alongside it. The ceiling arrives fast: macOS 14 and Apple Silicon are hard requirements, and the tool does not run on Windows, Linux, or Intel Macs.
Bottom line: Pick this when you need a private, cited audit trail across your own work data on Apple Silicon — skip it when your team works across operating systems or when you need any form of shared, collaborative memory.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Every answer carries numbered citations linked to the original source — email, Slack message, or file — so you can verify claims in seconds instead of hunting across apps where the context originally lived.
- Full-text and vector retrieval run without any AI model installed, which means search works immediately after connecting your sources and does not depend on a local model being available or correctly configured.
- Contradiction detection surfaces cases where two sources say different things and tracks them as open loops, so discrepancies in headcounts, dates, or decisions get flagged before they become day-of problems rather than after.
- The entire index is a single SQLite file in your home folder with no telemetry or server dependency, which means a subpoena, a data breach, or a vendor shutdown cannot expose your working history.
- Optional Ollama integration hot-plugs a local language model without reconfiguration, so you can switch between extractive search and conversational answers based on what you have installed — and the tool degrades gracefully when the model is off.
Cons
Sign in to edit- Hard platform requirements — macOS 14 and Apple Silicon — mean anyone on an Intel Mac, Windows machine, or Linux workstation cannot run the tool at all; teams with mixed operating environments will need a different solution from day one.
- The local AI models the vendor recommends for conversational answers require 16–32 GB of RAM; on machines at or below that ceiling, running indexing and a large local model simultaneously creates resource contention that affects both retrieval speed and model response time.
- There is no shared or collaborative memory layer — every user's index is isolated to their own Mac — so teams that need a common knowledge base or want to query each other's context will switch to a cloud-based RAG tool or a shared vector store instead.
- At v0.1.4 the connector set covers common work apps, but the absence of an API or plugin interface means sources not on the supported list — internal wikis, project management tools, CRMs — cannot be added without contributing to the open-source codebase or waiting for official support.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- macOS 14+ (Apple silicon)
- API Available
- No
- Self-Hosted
- Yes
- Last Updated
- 2026-06-19T00:18:17.352Z
Best For
Who it's for
- Mac users needing offline personal knowledge management
- Professionals who want verifiable citations from their own data
- Anyone avoiding cloud-based note or search tools
What it does well
- Find decisions and context from past emails or Slack threads
- Retrieve cited information from files and screenshots
- Maintain a private working memory across work apps
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Ultramemory
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Recommended skills for this tool
Auto-curated by the AIDiveForge recommendation matrix. These skills are predicted to enhance this tool based on category, capability, and domain signals.
-
Meeting Summary Template transform 32%
Turn a raw transcript into a decision-focused recap: outcomes, owners, deadlines, open threads.
Why: category partial · caps 0/0 · domain ops
-
Standup Note Synthesizer transform 32%
Merge individual standup bullets from multiple people into a single team digest with blockers surfaced to the top.
Why: category partial · caps 0/0 · domain ops
-
Runbook Skeleton post 32%
Produce a first-draft runbook from a postmortem — detection, diagnosis, mitigation, rollback — so the next incident has a template to follow.
Why: category partial · caps 0/0 · domain ops
-
OKR Draft Critiquer post 32%
Score draft OKRs against SMART criteria and the outcome-not-output rule, with suggested rewrites for each failing key result.
Why: category partial · caps 0/0 · domain ops
Frequently Asked Questions
- Is Ultramemory free?
- Yes — Ultramemory is fully free to use. There is no paid tier.
- Is Ultramemory open source?
- Yes. Ultramemory is open source.
- Can I self-host Ultramemory?
- Yes. Ultramemory supports self-hosting on your own infrastructure.
- What platforms does Ultramemory support?
- Ultramemory is available on: macOS 14+ (Apple silicon).
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Search across disconnected work apps produces fragments. Ultramemory is a Mac application that indexes email, Slack channels and DMs, Google Drive documents, calendar events, local folders, screenshots, audio transcripts, and Apple Notes into a single SQLite database stored in your home folder. You query it in plain language — ‘when did we decide to push the launch?’ — and receive a narrated answer where every sentence carries a numbered citation linked to the original source, displayed in chronological order of events.
The privacy architecture is the core differentiator. There is no account, no sync service, and the vendor states there is no telemetry — the database file lives at a standard macOS path, readable by you and nothing else. Every connector can be scoped or paused per channel, per folder, or per person. Export and full deletion are one-click operations. The AI layer is optional: Ollama integration is detected automatically, and if no model is running, the tool falls back to ranked search results rather than breaking.
Ultramemory fits solo professionals on Apple Silicon who need verifiable recall across their own work history — the kind of person who has been asked ‘why did we make that call?’ and had to reconstruct the answer from three browser tabs. It does not fit teams that need shared knowledge bases, Windows or Linux users, or anyone whose indexing needs exceed what a single Mac can process locally. The contradiction detection feature — which surfaces cases where two sources disagree and flags them as open loops — adds practical value in contexts like event planning or project management where decisions evolve across threads.
The tool requires macOS 14 or later and Apple Silicon. The AI models the vendor recommends — gemma4:26b for best answers and qwen3:4b for lower RAM — run through Ollama, installed separately via two terminal commands the docs provide. The underlying store is a standard SQLite file, which means it is inspectable and exportable without any proprietary tooling. The project is open-source and free with no paid tiers.