Real-time Analytics

Complete LLM Observability

Track every request, token, and dollar across all models and providers. Real-time dashboards with zero instrumentation.

01Spend

Where the money goes

Cost, caching and usage across every model and provider, current enough to act on.

Cost by Model7 DAYS

Cost

$1,247

Requests

143.2K

Tokens

48.2M

Avg latency

412ms

opus-4.6gpt-5.4gemini-3.1-prodeepseek-r3llama-4

02Performance

Latency, logs, sessions and per-agent behaviour, down to a single call.

PerformanceLIVE

Success rate

99.82%

P95 latency

445ms

Model health: success rate and latency per model
Model	Health	Latency	Success
opus-4.6	Healthy	892ms	99.9%
gpt-5.4	Healthy	445ms	99.2%
gemini-3.1-pro	Warning	512ms	98.8%
deepseek-r3	Healthy	678ms	99.5%

Request LogsLOGS

The seven most recent requests: model, status, latency, tokens and cost
Model	Status	Latency	Tokens	Cost
opus-4.6chain	200	1.2s	3,847	$0.042
gpt-5.4chain	200	890ms	2,104	$0.018
gemini-3.1-prochain	200	445ms	1,256	$0.008
deepseek-r3	429	12ms	0	$0.000
opus-4.6chain	200	2.1s	5,892	$0.064
gpt-5.4	200	678ms	1,893	$0.015
llama-4chain	500	340ms	0	$0.000

Showing 7 of 143,247

No credit card required.