Usage and cost estimate
Usage and cost estimate shows how much your organization consumes against plan quotas and what that usage likely costs in model API fees.
Where to find it
Section titled “Where to find it”- Log in to the dashboard.
- Open Dashboard → Usage (or Billing → Usage).
Metrics explained
Section titled “Metrics explained”| Metric | Description |
|---|---|
| Messages | User + assistant turns across widget and API deployments |
| Tokens | LLM input/output tokens driving model cost |
| Embeddings | Indexing and re-indexing charges from KB uploads and auto-retrain |
| Storage | Document and snapshot storage across knowledge bases |
| Estimated cost | Projected API spend based on current model list prices — not your final invoice |
Figures are estimates for operational visibility. Official charges appear on Invoices.
Usage vs plan
Section titled “Usage vs plan”The bar or percentage compares consumption to your tier limits:
| Resource | Typical quota |
|---|---|
| Messages per month | Varies by Free, Starter, Standard, Pro, Enterprise |
| Documents per KB | See Usage and quotas |
| Team seats | Admin + editors + viewers |
| Assistants | Count limits per plan |
When usage approaches 80–100%, consider Upgrade or reducing high-volume deployments.
Cost drivers
Section titled “Cost drivers”| Activity | Impact |
|---|---|
| High-traffic widget | Message and token volume |
| Large KB re-index | Embedding usage |
| Advanced models (Standard+) | Higher per-token cost — Model selection |
| Tool calls (Starter+) | Extra LLM rounds + external API latency |
Actions
Section titled “Actions”- Set deployment rate limits — Deployment rate limits
- Archive unused assistants — Clone, archive & delete
- Review analytics — Analytics tiers by plan