AI, OpenRouter and Pricing
LLM Cost (OpenRouter)
pinakea uses OpenRouter to access cloud AI models. This page explains where OpenRouter credit can be used and how to keep spending controlled.
Final measured pinakea timing and cost guidance is not published yet. Avoid treating this page as a cost calculator; use OpenRouter’s dashboard and pinakea’s in-app spend indicators for actual usage. For the in-app monthly counter to reflect pinakea alone, use a dedicated OpenRouter API key only with pinakea.
Where OpenRouter Credit Is Used
OpenRouter is used for:
- Embeddings for semantic search
- Automatic item summaries
- AI titles
- AI tags
- Chat
- Full item summaries
- Day/daypart summaries
pinakea sends provider.zdr=true with every OpenRouter content request, asking OpenRouter to use providers that advertise Zero Data Retention. This routing request is not an absolute provider guarantee. Chat may use reasoning; background jobs do not.
Zero Data Retention on embeddings is not an embedding migration. Embeddings still use qwen/qwen3-embedding-8b, so existing embeddings stay in the same vector space and do not need to be regenerated just because of this routing policy.
Two model choices per Set in Settings → AI. Background work defaults to a cheaper model; chat defaults to a faster one.
Model Price Reference
OpenRouter ZDR-listed base prices as of 1 July 2026:
| Model | Context | Input $/1M | Output $/1M | Relative to DeepSeek |
|---|---|---|---|---|
| DeepSeek V4 Flash | ~1M | $0.098 | $0.196 | Background default, lowest cost |
| Gemini 3.1 Flash Lite | ~1M | $0.25 | $1.50 | On-demand default; ~2.6x input / ~7.7x output |
| Grok 4.3 | 1M | $1.25 | $2.50 | ~12.8x input / output |
| GPT-5.4 | ~1M | $2.50 | $15.00 | ~25.5x input / ~76.5x output |
DeepSeek is by far the cheapest of these options at the rates above. Long-context or tiered billing can be higher for some models.
For current prices, check OpenRouter’s model list on their website.
Initial Library Processing
When you first add sources to pinakea, each processed item needs embeddings plus summaries, titles, and tags. This is usually the largest one-time OpenRouter usage event for a Set.
Actual usage depends on content length, provider pricing, and how much work is regenerated. Larger Sets naturally require more processing than smaller Sets. OpenRouter reports the real per-request cost in usage payloads, and pinakea uses that value when available.
Ongoing Usage
After initial processing, OpenRouter usage can come from:
- New items arriving through sources
- Chat conversations
- Search query embeddings
- Reprocessing after source changes or explicit regeneration
Cost Control
Set a spending limit on your OpenRouter key to cap your costs:
- Set a spending limit on your API key in OpenRouter’s dashboard.
- Use a dedicated API key only for pinakea, so OpenRouter key usage matches pinakea usage.
- Monitor spending in pinakea’s status bar and OpenRouter’s dashboard.
- Avoid automatic top-up unless you explicitly want it.
See OpenRouter and BYOK for setup instructions.