Cost intelligence
Know exactly what every call costs.
Foglamp prices every generateText and streamText call from real token usage — broken down by model, agent, and customer, in real time.
Spend · last 24h$842.17
$842
today
$25.3k
this month
98%
priced
Attribution
Every dollar, attributed.
Stop guessing where spend goes. Foglamp splits cost across models, agents, workflows, and even individual customers — using live provider pricing so the numbers actually match your bill.
- Per-model, per-agent, and per-customer breakdowns
- Live pricing for OpenAI, Anthropic, and 100+ models
- Cached vs. fresh token accounting
Spend by model
Regressions
Catch cost spikes before billing does.
A prompt change ships and spend doubles overnight. Foglamp surfaces the jump the moment it happens and ties it back to the exact agent and model driving it.
- Day-over-day deltas on every metric
- Drill from a spike straight to the responsible traces
- Pair with alerts for a hard daily spend ceiling
Cost over time
Your agents are running in the fog.
Cost, latency, errors, eval scores — all there, all invisible. Wrap your model and turn the light on.