Fleet Analytics

Total tokens · cost per inference · regional coverage

Updated 4:12:16 AM

Active Nodes

551

across all regions

Tokens / hr

85M

fleet-wide throughput

Req / sec

1,215.8

real-time aggregate

Cost / 1k tokens

$0.00105

weighted avg, all routes

Cloud Savings

$0.94k

vs cloud-only this hr

Success Rate

98.4%

fleet-wide p50

Click any two regions to inspect latency & transfer cost between peer nodes

Select 2 regions to inspect the link

Local: 53%

Mesh: 26%

Cloud: 21%

90-day history + 30-day forecast · adjust model offloading policy to see savings impact

Projected monthly (cloud-only)

$0.87

Monthly cost with Local First

$0.25

→

Monthly savings vs cloud-only

$0.62(71%)

Annual projection

$7 saved / yr

Region	Nodes	Tokens/hr	Req/sec	Cost/1k tok	p50 Latency	p99 Latency	Success %
US West	58	27M	128.4	$0.00091	114ms	216ms	99%
EU West	124	19.3M	154.9	$0.00136	54ms	109ms	99.8%
US East	81	15.56M	23.1	$0.00131	99ms	213ms	98.3%
EU Central	62	7.2M	148	$0.00126	141ms	321ms	97.5%
AP East	19	7.15M	26.6	$0.00061	77ms	258ms	99%
SA East	83	6.7M	353.7	$0.00084	153ms	343ms	96.7%
AP South	124	2.08M	381.1	$0.00109	51ms	176ms	98.2%

Data refreshes every 8 seconds in live mode. All figures are fleet-wide aggregates across all active peer nodes and cloud fallback routes.