🔋 How Energy Works
TreeOS operations like creating nodes, editing content, and running scripts cost energy. Your balance refills automatically once per day based on your plan.
💬 AI Chat
TreeOS requires you to connect your own LLM provider to use AI features. You bring your own API key and pay your provider directly for LLM usage.
Successful AI chat messages are free and do not cost energy. TreeOS operations triggered by the AI (creating nodes, editing notes, etc.) still cost energy as normal.
If an AI call fails (bad endpoint, invalid key, etc.), 2 energy is charged to prevent abuse.
📊 Energy Costs
Fixed Actions
Edit status, value, goal, name, schedule1
Move node, delete node, invite1
Prestige, run script, transaction2
Create node3
AI chat messagesFree
🎫 Plans
Your plan determines how much energy you receive each day. TreeOS offers three tiers:
Basic
350 / day
Free forever. Enough for daily note-taking, tree management, and AI chat. No file uploads.
Standard
1,500 / day
For active users. File uploads up to 1 GB, more room for scripts, and larger trees.
Premium
8,000 / day
For power users and teams. No file size limits and large-scale understanding runs.
Plans can be purchased or upgraded through your profile page via Stripe. If your plan expires, you're automatically moved back to Basic and your daily energy resets accordingly.
🔄 Daily Reset
Your daily energy refills to your plan's limit once every 24 hours, measured from your last reset. The reset replaces your daily balance and does not stack. Additional purchased energy is not affected by the daily reset.
🧠 LLM Connections
TreeOS requires you to connect your own LLM provider to use AI features like chat, understanding, dreams, and raw idea placement. You can use any OpenAI-compatible API endpoint. We recommend OpenRouter for the easiest setup. It gives you access to hundreds of models with one API key.
LLM Slots
TreeOS supports multiple LLM connections so you can use different models for different tasks. There are two levels of assignment:
Profile LLMs
Profile Chat (used everywhere unless overridden)
Raw Ideas (auto-placement of raw ideas)
Per-Tree LLMs
Placement (creating and organizing nodes)
Respond (conversation and chat)
Notes (writing and editing notes)
Understanding (compression and encoding)
Cleanup (expand and reorganize)
Drain (short-term memory placement)
Each tree can have its own LLM for each stage. If a tree slot isn't set, it falls back to the tree's placement LLM, then to your profile default. This lets you use a fast, cheap model for bulk operations like cleanup and a more capable one for conversation or understanding.
How to connect: When you first sign up, the setup page will walk you through connecting your first LLM. You can manage connections and slot assignments anytime from your profile's energy page. Any OpenAI-compatible endpoint works. Just provide a base URL, API key, and model name.