Available on: All plans | Role: Admin+ | Status: Generally Available
Overview
The Performance & Cost dashboard gives you operational visibility into how your AI models are performing and what they cost. Filter by teamspace to drill into specific workspaces.
Four key performance indicators are displayed at the top:
| Metric | Description |
|---|
| Query Success Rate | Percentage of queries that returned a successful response |
| Avg Response Time | Average time for the AI to generate a response |
| Token Usage | Total tokens consumed across all models |
| Error Rate | Percentage of queries that resulted in errors |
Cost Breakdown
A dedicated cost section shows financial impact:
| Stat | Description |
|---|
| Total Monthly Cost | Aggregate cost of all AI model usage this month |
| Cost per Query | Average cost for a single AI interaction |
| Cost per User | Average cost divided by active users |
| Total Queries | Number of queries processed this month |
- Model name and provider
- Performance category — Excellent, Good, or Average (color-coded)
- Individual metrics — Response time, success rate, and usage volume
Teamspace Filter
Use the teamspace filter dropdown at the top to scope metrics to specific teamspaces:
- All Teamspaces — Aggregate view across the organization
- Single teamspace — Metrics for one specific workspace
- Multiple teamspaces — Combined metrics for selected workspaces
Alerts
The system generates alerts for anomalies like:
- High error rates on specific models
- Unusual response time spikes
- Token usage approaching limits
Alerts are color-coded by severity (high = red, medium = amber, low = green).
Charts
| Chart | Description |
|---|
| Response Time Trends | Average and P95 response times over time |
| Cost Distribution | Breakdown of costs across models |
| Token Distribution | Token usage by model |
| Request Types | Query volume by type |
