”How much have I spent this week and what’s my balance?”
The agent checks billing:billing_balance— returns current credit balancebilling_usage— returns spending breakdown for the last 7 days
Current balance: $487.30 Last 7 days spending: $112.70Your H100 inference server accounts for 75% of your spend. It’s been running 24/7. At this rate, your balance covers another ~30 days.
”Find anything that’s running but not being used and shut it down.”
The agent cross-references instance status with GPU utilization:list_instances— gets all running instancesssh_execute— checksnvidia-smion each running instancebilling_usage— calculates cost per instance
Checked 4 running instances:After you confirm:dev-box and test-bench have 0% GPU utilization. Together they’re burning $0.70/hr. Want me to terminate both?
delete_instance— terminatesdev-boxdelete_instance— terminatestest-bench
Terminated dev-box and test-bench. Saving 16.80/day).
”Give me a monthly projection. Can I afford to run an H100 for the rest of the month?”
The agent calculates:billing_balance— checks current balancebilling_usage— gets daily burn rate from the last 30 dayslist_instances— gets current active instances and their per-hour cost
Balance: 2.50/hr (inference-prod H100) + 4.10/hr** At $4.10/hr, you have ~119 hours (~5 days) of runway. If you shut down the A100 after training completes (~6 more hours), the H100 alone burns $2.50/hr — that gives you ~192 hours (~8 days). To run the H100 for the remaining 12 days of the month, you’d need ~233 short**. Options:
- Add $250 in credits from the dashboard
- Switch to an A100 for inference ($1.60/hr) — covers the full month