Usage & Quality

demo

Aggregated usage, cost estimates, quality metrics, and provider readiness. All data shown is demo/sample — production metering is not connected yet.

Usage Overview

demoTotal Sessions28demo this period
demoConversation Minutes1m 4sacross all agents
demoRendered Video Minutes0Studio not yet active
demoActive Agents1Ethen Guide (published)

Cost Estimate

demoSTT$0.04
demoTTS$0.0075
demoLLM$0.0039
demoWebRTC$0.0000
demoGPU$0.0000
demoTotal$0.27per session avg

Sessions

demo
ServiceUsageDemo RateEst. Cost
STT45s$0.0059/min (demo estimate)$0.04
TTS500 chars$0.015 per 1k chars (demo estimate)$0.0075
LLM120 in / 450 out$0.00015 input / $0.0006 output per 1k tokens (demo estimate)$0.0039
WebRTC0 min$0.004/min (demo estimate)$0.0000
GPU (future)0 sec$0.0013/sec (demo estimate)$0.0000
Storage0 B$0.023/GB-month (demo estimate)$0.0000

Quality

demoAvg Latency (P50)1.1send to first audio
demoP95 Latency2.1send to first audio
demoFallback Rate0/20 of 2 turns
demoInterruptions1per session
demoTTS Success100%demo sample
demoSTT Success100%demo sample
demoLLM Success100%demo sample
demoErrors0demo sample

Provider Readiness

STTdeepgram
active
TTSelevenlabs
active
LLMstatic
active
WebRTCnone
not_configured

Agent Performance

Sessions

demo
AgentTurnsAvg LatencyEst. CostStatus
Ethen Guide3990ms$0.07active
Website Concierge21.1s$0.05draft
Support Agent———template
Training Coach———template

Limitations

  • • Production metering is not connected yet. All usage and cost data shown is demo/sample.
  • • Provider pricing uses placeholder estimates from lib/avatar-cloud/metrics.ts.
  • • Rendered video minutes, WebRTC session time, and GPU seconds are zero — these services are not active.
  • • No database, auth, Stripe, or billing enforcement active.
  • • Agent performance data for Support Agent and Training Coach is placeholder — these agents are templates only.