Model usage

Every request is itemised, so a surprise on the bill always traces back to a specific call.

Sample data. Nothing on this page is connected to an account. Every figure, row and status below is placeholder content.

Calls, 28 days

170K

Cost, 28 days

$166.13

Median latency

0.34s

across all routes

Fallback rate

2.1%

primary route healthy

Calls per day

last 28 days
13579111315171921232527
Counts in thousands. Placeholder series.

By model

ModelCallsTokens inTokens outCost
atlas-5.6-solo41K92M18M$55.74
swift-3.8-flash96K211M31M$29.72
lumen-5-pro6.4K19M4.1M$52.09
delta-v4-flash23K47M9.8M$25.28
quill-4.63.9K8.4M2.2M$3.30

Request log

TimeKeyModelRouteLatencyTokensCost
21:04:19productionatlas-5.6-soloprimary0.41s1,204$0.0026
21:04:17batch-jobsswift-3.8-flashprimary0.22s3,880$0.0009
21:04:12productionlumen-5-profallback1.08s2,415$0.0184
21:04:09stagingswift-3.8-flashprimary0.19s612$0.0001
21:04:02productionatlas-5.6-soloprimary0429
21:03:58batch-jobsdelta-v4-flashprimary0.34s5,102$0.0022
21:03:51productionquill-4.6primary0.56s1,880$0.0008