AI, OpenRouter and Pricing
Leveraging AI
A Precision-Engineered Pipeline for Your Personal Knowledge
pinakea does not just “use AI.” It orchestrates a sequence of specialized stages so raw items become structured knowledge you can search, summarize, and chat with.
This page explains the pipeline architecture and why each stage exists.
Online AI through OpenRouter
All model-based stages — embeddings, summaries, titles, tags, and chat — use cloud models through OpenRouter with your own API key.
- Uses cloud models for embeddings and pipeline generation.
- Supports parallel processing for large imports.
- Requires internet access and a valid OpenRouter API key.
- Keeps provider billing and spend limits directly in your OpenRouter account.
The AI Pipeline: From Raw Content to Refined Knowledge
When you add sources (folders, clips, mail, notes), each item flows through a staged pipeline.
Stage 1: Content Extraction
pinakea first normalizes content into clean text:
- Markdown/text: direct parse.
- PDFs: native extraction, with OCR fallback for scanned pages.
- Images/screenshots: OCR via Apple Vision.
- YouTube clips: transcript retrieval so spoken content becomes searchable.
- Web pages and HTML files: Full capture stores the page text as-is, with an optional Clean view in the Original pane for hiding obvious page chrome when reading.
Stage 2: Chunking
Long content is split into overlapping chunks so semantic retrieval remains accurate across boundaries.
- Typical chunk size is tuned for retrieval efficiency.
- Overlap preserves context between adjacent chunks.
Stage 3: Embedding
Each chunk is converted into a semantic vector.
| Model | Details |
|---|---|
| Qwen3 Embedding 8B | Cloud, high-dimensional semantic retrieval through OpenRouter |
This is what allows meaning-based search (“delivery schedule” matching “project deadlines”).
Stage 4: Automatic Summary
Each item receives an automatic summary for fast scanning.
| Model | Details |
|---|---|
| Background AI processing (DeepSeek by default) | Cloud, high throughput through OpenRouter |
Stage 5: Intelligent Title
pinakea generates better descriptive titles from content, so your timeline remains readable at scale.
Stage 6: Automatic Tags
pinakea generates conceptual tags (not simple keyword extraction) to improve browse and retrieval flows.
Why These Model Choices
Embeddings: Retrieval Quality First
Embedding quality determines search and chat grounding quality. pinakea prioritizes:
- semantic depth over lexical matching,
- robust retrieval at large library sizes,
- consistency within the embedding vector space.
Pipeline Generation: Speed + Consistency
For automatic summaries/titles/tags, pinakea optimizes for:
- throughput during ingest,
- stable output quality,
- predictable cost behavior across long-running pipelines.
In Settings → AI each Set has two models: Background AI processing (default DeepSeek) for indexing work, and On-demand AI processing (default Gemini) for chat and summaries you open. Embeddings use Qwen3.
The Clean view for web clips and HTML files is different from background item processing: it runs only when you choose Clean in the Original pane’s Full | Clean control and no cached Clean view exists. It uses DeepSeek V4 Flash through OpenRouter to hide obvious page chrome while keeping your stored Full capture unchanged. New uncached Clean generation is free while monthly processing remains, pauses at Free monthly exhaustion, and existing cached Clean views remain readable.
Chat: Conversations with Your Knowledge
Search gets you to relevant items. Chat synthesizes across them.
Retrieval Flow
- Your question is embedded into the same semantic space as your indexed chunks.
- Retrieval selects the most relevant chunks by semantic proximity.
- Multi-turn context keeps seed + incremental evidence aligned.
- Citations tie generated claims back to source items.
Citation Guarantees
Answers include clickable references so you can jump directly to supporting source content.
Summaries and Chat
Automatic Summaries
- Precomputed during pipeline processing.
- Optimized for quick scanning in timeline workflows.
- Free on every plan.
On-Demand Summaries
On-demand summaries are day, daypart, and tag summaries. They are available on Free, Pro Monthly, and Pro Lifetime. They do not spend the Free monthly processing allowance while processing remains, but new generation pauses when the monthly processing limit is exhausted.
Chat Prompts
Chat is the synthesis surface for follow-up questions over an item, selection, search result, day, daypart, or current timeline context. Chat prompts do not spend the Free monthly processing allowance.
Privacy and Data Flow
- Source files and the pinakea library (database, index, generated results) are stored on your Mac.
- Items you add are processed automatically through OpenRouter. The content needed for each request goes from your Mac to OpenRouter to the selected provider over encrypted connections.
- pinakea sends
provider.zdr=truewith every OpenRouter content request, asking OpenRouter to route it to providers that advertise Zero Data Retention. This request is not an absolute provider guarantee.
Cost Control
Cloud usage is BYOK through OpenRouter:
- you control spend limits,
- you see spend in OpenRouter + in-app status reporting,
- no pinakea markup on model usage.
Detailed cost examples: LLM Cost (OpenRouter).