Total requests ↗
0No requests yetAGENT OBSERVABILITY
Every inference, accounted for.
A clear view of your routing, reliability, and spend.
Connect your agent
Use your OmniDeploy key to view real request history.
Need an agent key?
Save this key securely. It is shown only once.
Inference spend $
$0.00Based on reported token usageEstimated savings ✧
$0.00Compared with the priciest eligible routeProvider attempts ⇄
0Includes retries and fallback attemptsProvider activity
Where your agent's requests are going
Your first request starts the story. Send an inference to see its provider, latency, and cost here.
Routing that learns.
Each agent has its own history. Recent reliability and latency influence which provider handles the next request.
Learning window50 attempts
Success threshold15 percentage points
Latency threshold30% above best
Recent requests
Latest 12 inference attempts
| Provider / Model | Result | Latency | Cost | Time |
|---|
Savings use actual input and output tokens priced against the most expensive configured provider for the same model. Prices are configurable snapshots, not invoices.