Monitor LLM costs, infrastructure usage, and per-session breakdowns.

Usage Tracking

Workspace provides detailed usage tracking so you always know where your credits go.

Usage Dashboard

Go to Billing > Usage to see detailed session history. Each session represents one period of a team running.

Session Details

Each session shows:

• Team name — which team was running
• Duration — how long it ran
• LLM cost — total spend on AI model tokens
• Infrastructure cost — compute cost for the running environment
• Total cost — combined spend
• Token counts — input and output tokens used

Filtering

Filter sessions by time range:

• Last 7 days
• Last 30 days
• Last 90 days

Cost Breakdown

LLM Costs

Each AI model has different per-token pricing. Costs depend on:

• Which model the persona uses
• How many tokens are generated (input + output)
• How complex the conversations are

More capable models (Claude Opus, GPT-4) cost more per token than smaller models (Gemini Flash Lite, Llama).

Infrastructure Costs

A flat rate of $0.02/hour per running team covers compute, persistent storage, and network.

Live Sessions

Currently running teams show a Live badge in the usage view. Their costs are estimated in real-time and finalized when the team is stopped.

Per-Company Tracking

Usage is tracked per company. Switch between companies on the billing page to see each company's usage separately.