Transparent Infrastructure Pricing
Bypass per-minute API premiums. Pay for flat GPU compute capacity and carrier rates with zero hidden markups.
No Hidden Markups
Three Pricing Elements
We bill platform operations transparently. Bypassing variable AI service markups lets you pay directly for access, compute, and carrier routes.
Platform Fee
$99/mo FlatUnlocks admin dashboards, multi-tenant lead isolated databases, sliding-window dial session controllers, and immediate CRM sync queues.
Raw Compute Cost
GPU Instance Cost or BYO APIPay for GPU node time which processes concurrent calls. Or choose API Mode and connect your own provider keys.
Telephony Carrier Cost
~$0.005/min Carrier RateRouted directly using standard SIP trunk providers. Valyrian Voice does not add markups to telephony carrier usage rates.
Core Pricing Principle: Valyrian Voice does not monetize AI models in GPU mode.
Outbound Calling Cost Calculator
Estimate how many SIP minutes your outbound campaign will actually consume—not just how many calls you make. Configure answer rates, voicemail, conversation lengths, and carrier billing increments to get a realistic estimate of your monthly calling cost.
Campaign Volume
Set monthly call volume directly, or estimate using daily volume × working days.
Call Outcome Distribution
Configure expected outcome percentages across your campaign calls.
Average Call Durations
Configure average actual duration per connected category. Calculated internally in seconds.
Estimated SIP Costs & Minute Summary
Call Outcome & SIP Billing Breakdown
Detailed breakdown of calls and per-call billing increment rounding across categories.
| Outcome Category | Calls (% Distribution) | Avg Actual Duration | Billable Min / Call | Total Billable Min |
|---|---|---|---|---|
| No answer / busy / unreachable | 350 (35%) | — | 0 min | 0 min |
| Declined / rejected before connection | 100 (10%) | — | 0 min | 0 min |
| Voicemail / answering machine | 150 (15%) | 30 sec | 1 min | 150 min |
| Very short human conversation | 100 (10%) | 45 sec | 1 min | 100 min |
| Normal human conversation | 200 (20%) | 4 min | 4 min | 800 min |
| Long human conversation | 100 (10%) | 9 min | 9 min | 900 min |
| Total | 1,000 calls | — | — | 1,950 min |
Call Outcome Distribution
1,000 attemptsBillable Minutes Breakdown
1,950 total minEstimated Concurrent Call Capacity
Based on 1,000 monthly attempts
Across 550 connected calls
1 GPU node handles up to 100 concurrent streams
* Estimated concurrent call capacity provides an approximation for trunk channel provisioning and GPU instance allocation. Actual concurrency varies with dialer pacing algorithms and answer rate spikes.
* Actual answer rates, voicemail pickup rates, and conversation lengths vary by campaign, geography, list quality, dialing strategy, and audience. Adjust the assumptions above for your specific campaign.
Ready to start calling?
Configure your ValyrianVoice deployment based on your estimated calling volume and eliminate per-minute AI API premiums.
Interactive Campaign Calculator
Estimate and compare campaign costs across different call volumes.
Select Volume & Duration
Adjust sliders to fit your anticipated campaign size.
Platform Access
$99/month flat fee for CRM integrations and workspaces.
Compute Cost
GPU node compute billed pay-as-you-go from provider.
Carrier SIP
VoIP SIP trunk lines direct-billed at raw cost (~$0.005/min).
* The Platform Access Fee ($99/mo) is billed flat per organization and is not factored into the campaign cost calculations.
Campaign Cost Projections
Calculations prove that compute-based models yield significant advantages over per-minute API variable pricing tiers as campaign size increases.
Why Traditional Pricing Fails at Scale
API voice systems charge a linear markup for every model component (transcription engines, reasoning models, and voice synthesizers) on a per-minute basis. Because costs grow in direct proportion to your dialing minutes, campaigns quickly become unprofitable as your scale expands.
Valyrian Voice runs stateless voice engines on dedicated bare-metal GPU clusters. By packing up to 100 concurrent calls per GPU node at a flat hourly cost, we remove variable margins. This makes outbound and inbound operations significantly cheaper as call volume grows.
Component Cost Comparison Table
Review how Valyrian Voice compares to traditional API integrations on a layer-by-layer level.
| Cost Category | Valyrian Voice (GPU Mode) | VAPI | Bolna.ai |
|---|---|---|---|
| Platform Fee | $99 / month flat | $0 | $0 |
| Platform / Middleware Markup | $0.00 / min | $0.05 / min | $0.02 / min |
| Speech-to-Text (ASR) | Included in GPU compute | $0.01 / min (Deepgram) | Included in flat rate |
| Language Reasoning (LLM) | Included in GPU compute | $0.004 / min (GPT-4o mini) | Included in flat rate |
| Text-to-Speech (TTS) | Included in GPU compute | $0.036 / min (ElevenLabs) | Included in flat rate |
| Telephony SIP Carrier Trunk | ~$0.005 / min (Direct billed) | Included in platform fee | Included in flat rate |
| Total Billing Model | Flat GPU hours + Carrier | $0.10 / min flat rate | $0.05 / min flat rate |
Ready to Scale Your Voice Operations?
Launch outbound campaigns, deploy inbound AI receptionists, and cut your calling costs, all from one platform.