LLMnesia vs Readwise for AI Conversation Capture

What Readwise does

Readwise is a reading retention tool. It syncs highlights from Kindle, Instapaper, Pocket, articles, PDFs, and other reading sources, and surfaces them through spaced repetition to help you remember what you've read. Readwise Reader is its companion read-later app.

Readwise's value is in the reading pipeline: capturing what you encounter, making sure you actually remember it, and building a searchable library of your reading highlights over time.

For AI-adjacent research workflows: Some users import articles and papers into Readwise Reader before using them as context for AI conversations. The article stays in Readwise; the AI conversation about that article is separate.

What LLMnesia does differently

LLMnesia is not a reading tool. It's a conversation retrieval tool. It indexes the sessions you have with AI tools — the questions you asked, the answers you received, the code generated, the analysis produced — and makes them searchable.

Where Readwise captures your input (what you read), LLMnesia captures your output (what you got from AI). These are complementary, not competing.

The content type mismatch

The reason Readwise doesn't naturally extend to AI conversations is a content type mismatch.

Readwise handles documents: static content from a URL or file that you highlight specific passages from. Good document highlights are short, extracted from a known source, and manually selected for quality.

AI conversations are sessions: dynamic, multi-turn exchanges that produce value through the entire exchange, not through specific passages. The value is often in how the conversation evolved, what questions led to which answers, and the context around a specific response — not in a highlighted sentence.

Trying to use Readwise for AI conversations requires you to decide which parts to highlight at the time of the conversation, which reintroduces the discipline problem that LLMnesia solves by capturing everything automatically.

Where they overlap for researchers

Research-heavy users who use both tools often have a workflow like:

1. Encounter a relevant article or paper
2. Save it to Readwise Reader for reading and highlighting
3. Use an AI tool (Claude, Perplexity) to summarise, analyse, or extend the research
4. The article highlights live in Readwise; the AI conversation lives in LLMnesia's index
5. When revisiting a research question, search both: what did you read, and what did AI help you conclude

For this workflow, both tools are additive.

Choosing between them

Use Readwise if:

• Your primary information problem is reading retention and highlight capture
• You read heavily from web articles, books, or PDFs
• You want spaced repetition review of what you've read

Use LLMnesia if:

• Your primary information problem is finding past AI conversation output
• You use AI tools heavily for research, writing, or development
• You want cross-platform search across ChatGPT, Claude, Gemini, and others

Use both if:

• You're a researcher or knowledge worker who both reads heavily and uses AI tools heavily
• Your reading and AI workflows are intertwined — you read to inform AI queries, and AI output informs what you read next