Dashboards · 6 widgets · last 30 days
Overview
- Runs
- –
- Failure rate
- –
- Typical run
- –
- Slowest 5%
- –
- Spend
- –
Traces over time
How many runs your agent made, with failed runs marked in red.
Failure rate
The share of runs that hit an error, and how it moves.
Cost by model
Which models take the biggest share of the bill.
Response time
How long a run takes: the typical one (median) and the slow ones (95th percentile).
Cost per day
What your agent spent each day on models and tools.
Top users by cost
The end users who cost the most in this period.