Company profile · 3 min read
Recall: A personal knowledge base with API and MCP access
Recall saves and summarizes articles, videos, podcasts, PDFs and notes, then connects them into a searchable personal AI knowledge base.
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
Published · Updated
Recall is a personal AI knowledge base for saving, summarizing and revisiting articles, videos, podcasts, PDFs and notes. It fits a knowledge worker, researcher or lifelong learner who wants a curated library that can answer questions later instead of another stream of disposable summaries.
What it does
Recall’s current site describes a workflow that saves online content, summarizes it, organizes it into a knowledge graph and lets the user chat with the resulting library or the web. The product also includes notes, listening mode, quizzes and spaced repetition. API and MCP access are available on the public Free plan, so the knowledge base can feed the AI tools a user already works with.
The product’s differentiator is the archive around the summary. A user can save a podcast, video, article or PDF, add a note, connect it to related material and later ask across the collection. That makes Recall more useful for a founder or researcher who repeatedly returns to a body of sources than for someone who only needs a one-off answer.
Pricing is public. The pricing page lists Free at $0 with 10 AI summaries per month, Plus at $10 per month billed yearly with unlimited summaries, automatic organization, chat, listening mode, knowledge-graph connections, quizzes and bulk import, and Max at $38 per month billed yearly with model choice and bulk AI actions. The page says the Free plan includes API and MCP access. That makes the low-cost integration path unusually clear, though heavy users should understand summary and model limits before moving a research workflow into it.
The current homepage reports more than 700,000 professionals and displays a 4.7/5 rating based on 464 reviews. Those are company-published current-site claims, not an independent usage or review audit. No founder background is exposed in the retained sources, so I am not adding one.
What could make it the wrong choice
Recall depends on the user choosing what to save and trusting the system’s summaries and connections. It is not a source-of-truth system for a company’s regulated knowledge unless the team verifies retention, permissions, export and provenance requirements. The useful test is whether the saved library actually improves retrieval and learning over the user’s existing notes.
Editorial take
I would shortlist Recall for a founder or researcher building a personal source library and willing to curate it. The Plus plan is easy to understand, while API/MCP access makes the Free plan useful for experiments. If the problem is enterprise-wide knowledge governance rather than personal context, choose a system with stronger administrative controls.
Quick facts
| Field | Sourced detail |
|---|---|
| Product | Personal AI knowledge base, summaries, notes, chat and knowledge graph |
| Buyers | Researchers, founders, analysts, students and lifelong learners |
| Public pricing | Free; Plus $10/month yearly; Max $38/month yearly |
| Integrations | API and MCP access listed on Free; content includes articles, video, podcasts and PDFs |
| Main gate | Source curation, summary trust, export, privacy and personal-versus-team fit |
Sources checked
| Source | Checked |
|---|---|
| LAUNCH portfolio record | 2026-09-19 |
| Recall homepage | 2026-09-19 |
| Recall pricing | 2026-09-19 |
About the author
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.
First1000 ↗ · X ↗