voice-cost-clock

Live $/min clock for production voice-agent APIs

Each row is a real fetch against the provider's public pricing page. Refresh runs every 6 hours via node-cron. No mock numbers — rows that can't be parsed are shown as unparseable.

Leaderboard

…

# Provider Pricing model Base $/min Telephony $/leg $/10-min call TCO @ scenario Last fetch
Loading providers…
Unparseable rows (0)

    Recent price changes

    • No changes detected yet. First diff appears after the second snapshot.

    Custom calculator

    Methodology & data sources

    Every row is the result of a server-side fetch() against the provider's public pricing page (or a documented fallback URL). Extracted prices are normalized to USD per minute of conversation. Where a provider doesn't bundle telephony, we apply a documented baseline assumption (Twilio standard, $0.013/leg) — this is disclosed inside each provider's notes field. Token-billed providers (OpenAI Realtime) are converted to per-minute using an explicit audio-token assumption documented in their fetcher. See each row's expanded panel for the raw snippet and source URL.

    This is a price-watch, not a recommendation. Provider quality, latency, language coverage, compliance posture, and integration burden vary widely — compare those independently. We don't run pricing simulations against partner-provider rate cards.