Appearance
FeynmanLM vs Recall
Recall (recall.it, formerly getrecall.ai) is a self-organizing AI knowledge base: save articles, YouTube videos, PDFs, and podcasts; get AI summaries, auto-tagging, a knowledge graph, and spaced-repetition quizzes. It's the closest competitor to FeynmanLM in spirit — both are local-first, both cover many source types, both take retention seriously. The differences are in pricing model, platform, and how deep the review goes.
At a glance
| FeynmanLM | Recall | |
|---|---|---|
| Price | Free to download — no tiers, no purchase, no subscription | Free tier (10 AI summaries/mo); Plus $120/yr; Max $456/yr (billed yearly) |
| AI costs | Pay-as-you-go from a prepaid balance ($5 free to start): any model in Chat at provider API price +15%, or 1¢ per tool call via an AI subscription you already pay for | Included in subscription; model auto-selected on Plus, model choice only on Max |
| Platforms | Native macOS app (macOS 14+) | Web app + browser extensions, iOS, Android — no native desktop app |
| Data storage | Local SQLite + files in your private iCloud container, CloudKit sync | Local-first on device; sync via Recall's cloud (Google Cloud, EU) |
| Source ingestion | Automatic: Safari reading list, Apple Podcasts follows, PDFs, papers, books, YouTube, X posts | Manual saves via extension/app: articles, YouTube, PDFs, podcasts, X, Reddit |
| Review method | Feynman-technique dialogue with an AI tutor that reads the full source | Spaced-repetition quizzes (question/answer style) — paid tier only |
| MCP / assistant access | Full MCP control surface: search, schedule, run and save review sessions | MCP + API access (retrieval) |
| Export | Files in your iCloud Drive; SQLite you can open | Markdown zip export (web app only), API |
Pricing verified July 2026 — check recall.it/pricing for current numbers; Recall raised Plus pricing at its 2.0 relaunch in April 2026.
Pricing: $120+/yr subscription vs free app + pay-as-you-go
Recall's free tier stores unlimited content but allows only 10 AI summaries a month — the product's core is effectively paid. Plus is $120/yr billed yearly, and choosing which AI model answers you costs $456/yr on Max.
FeynmanLM has no tiers to climb, because it has no tiers: the app is free to download, every model is available immediately, and you pay only for the AI you run. That comes out of a prepaid balance topped up through Stripe, with $5 free to start. Use the in-app Chat with any model and you pay that provider's own API list price plus a 15% margin, metered on the exact cost the provider reports. Or connect whichever of Claude, ChatGPT, Gemini, or Grok you already subscribe to over MCP and pay 1¢ per tool call — your subscription covers the inference, the cent covers reading and writing your library, and a daily spend cap bounds runaway usage. The in-app ledger accounts for it to the cent. There's no tier where the model picker lives, because the model picker was never behind one.
Where Recall is better
- Works everywhere. Web app, iOS, and Android — save and review from any device. FeynmanLM is Mac-only.
- AI that stays free. Both apps are free to install, but Recall's free tier keeps giving you 10 AI summaries a month indefinitely at $0. FeynmanLM includes $5 of AI to start; past that, review draws on a balance you top up.
- Knowledge graph and auto-tagging. Recall automatically links related saves into a browsable graph; FeynmanLM organizes by source type, Projects, and Schedule instead.
- Augmented browsing. Recall's extension surfaces your saved knowledge while you browse the web.
- Reddit imports. Recall ingests Reddit threads; FeynmanLM doesn't.
Where FeynmanLM is better
- Deeper review. Recall's spaced-repetition quizzes are generated Q&A cards. FeynmanLM runs a Feynman-technique conversation: you explain the source in your own words and an AI tutor with the full text probes for gaps and follows up. And review isn't paywalled behind a higher tier — see Why Quizzing?
- Sources arrive by themselves. Recall ingestion is save-by-save via the extension. FeynmanLM auto-discovers from your Safari reading list, Apple Podcasts follows, iCloud folders, and reference managers — your library reflects what you actually consume.
- Podcasts as first-class sources. Follow shows, get full episode transcripts, read or listen, and get quizzed on them — not just saved links.
- Your cloud, not theirs. Both apps are local-first, but Recall syncs through Recall's servers on Google Cloud; FeynmanLM syncs through your own iCloud account, and the app's servers never store your library.
- A study planner. The weekly Schedule assigns sources to days with completion tracking and review history; Recall has a review queue but no planning surface.
- No tiers and no caps. No monthly summary allowance to ration, no model-choice upsell, no annual fee ticking in the months you don't study — you're billed on what you actually used, and nothing when you don't.
Which should you choose?
Choose Recall if you need Android, Windows, or the web — or you want a free place to pile up saves with light AI summaries and are happy with card-style quizzing at $120/yr.
Choose FeynmanLM if you're on a Mac and want the deeper loop: automatic source capture including full podcast transcripts, a weekly plan, Feynman-style review with the AI of your choice, and a library that lives in your own iCloud — for nothing up front and only what your AI use actually costs.