An AI panel debates your goal; a referee rules on re-checkable evidence
Build human-like AI voice agents with no-code or code, on carrier-grade telephony
Stage
Not given
Established
Founded
Not given
2011
Based in
Not given
US
Pricing model
Freemium
Freemium
Pricing
Free: $1.00 starting credit once. Credit plans from $10 top-up with no expiration. Your own key: 5% infrastructure fee after the first $5.00 of usage.
Pay as you go: Voice AI Agents $0.03/min (excluding telephony). SIP Trunking/Audio Streaming from $0.0010/min inbound, $0.0028/min outbound. WhatsApp messages from $0.0034/message. Enterprise plan starts at $1,000/month with custom pricing.
Key features
Panel of up to six AI models from different vendors
Referee rules on re-checkable evidence only
Searches academic databases: arXiv, OpenAlex, Crossref, Europe PMC
Executes Python code to verify claims
Customizable acceptance standards
Matches pause and resume without losing work
Shared ledger prevents re-derivation
Spending limit per match
No-code AI agent builder with visual drag-and-drop interface
Sub-500ms latency for natural real-time conversations
Support for 50+ languages and accents
Natural turn-taking with interruption detection
Self-improving agents with auto simulations and evals
Enterprise security with HIPAA, GDPR, SOC 2, PCI DSS compliance
Carrier-grade SIP trunking and audio streaming
AI noise cancellation and natural prosody
Data residency options in US, EU, or APAC
Integration with LiveKit and Pipecat frameworks
What makes it different
Referee decides on evidence, not argument quality
Panel from different vendors fails in different directions
Stuck is a targeted question, not model thrashing
Cost grows with ledger, not round count
Published matches become research record
Human-like voice quality with sub-500ms latency and natural conversation flow
Flexible architecture supporting both no-code and code-first development
Transparent pay-as-you-go pricing with $10 free credits to start