Token Usage & Shared Pool

Your subscription tier allocates a collective monthly token allowance shared across all agents in your workspace:

• A high-volume customer support agent answering hundreds of daily inquiries shares the same pool as a finance agent performing deep weekly reconciliations.
• When pool consumption reaches 80% and 95%, administrators receive automated in-app and email threshold alerts.
• If tokens reach 100%, non-critical scheduled routines pause until an instant top-up pack is purchased or the pool resets on your monthly billing date.
• Tokens do not roll over between billing cycles; your pool completely refreshes at each cycle start.

Not every business task demands maximum reasoning horsepower:

• Standard Models (1x Weight): Customer inquiries, appointment scheduling, meeting summarization, and rapid document queries.
• Genius / Advanced Reasoning Models (3x–5x Weight): Deep multi-table database queries, legal document contract analysis, multi-agent synthesis, and autonomous coding.

Botonom's Smart Router automatically selects the most efficient model tier based on task complexity, ensuring your token pool lasts longer.

1. Precise Directives: Write concise, clear directives rather than repetitive paragraphs.
2. Selective Knowledge Ingestion: Upload relevant chapters or documents rather than raw, uncompressed 500-page archives.
3. Task Scoping: Break massive multi-department workflows into modular steps with human verification gates.