Smallest AI
Pulse and Lightning — fixed-price low-latency voice for call centres and telephony.
Smallest AI focuses on the Pulse and Lightning series, providing ultra-low-latency text-to-speech starting at $0.05 per minute. Their platform is purpose-built for high-concurrency telephony systems and call centres, where clear, fixed PAYG pricing is preferred over complex credit models. End-to-end latency stays under 300ms (model latency under 100ms) — fast enough for interactive use, priced for high-volume contact centres.
Models
Pulse
TelephonySub-100ms model latency for live call agents
Input Price
$0.05 / minute
Input Types
Text
Output Types
Audio
Lightning
Built for high-concurrency PSTN / VoIP traffic
Input Price
$0.05 / minute
Input Types
Text
Output Types
Audio
Use Cases
Contact Centre Automation
Outbound and inbound call automation with per-minute pricing that fits contact-centre unit economics.
Telephony / IVR
PSTN and VoIP bot flows that need predictable latency and predictable cost per call.
High-Concurrency Voice
Stream thousands of concurrent voice sessions with fixed pricing — no surprise cost spikes during volume bursts.
Cost-Sensitive Voice Agents
Voice surfaces where ElevenLabs quality is overkill and Cartesia's premium isn't justified.
Why use Smallest AI on Nagent?
Nagent adds enterprise orchestration, observability, and workflow automation on top of Smallest AI's raw model capabilities.
Fixed-price PAYG fits contact-centre cost models cleanly
High-concurrency built into our voice-agent runtime — handle volume spikes without re-architecting
Drop-in replacement for ElevenLabs / Cartesia where quality budget is constrained
Direct integration with Twilio / Plivo / Telnyx via our telephony connectors
How to access Smallest AI on Nagent
Open Agent Studio
Navigate to Agent Studio in your Nagent workspace.
Select Smallest AI
Choose Pulse or Lightning under Audio / Voice in Model Configuration.
Connect Telephony Provider
Wire your Twilio / Plivo / Telnyx number to the agent — voice flow is live in minutes.
Common questions about Smallest AI
Real buyer and developer questions, answered. Click any item to expand.
Why pick Smallest AI over Cartesia for voice?
Pricing model. Smallest AI's flat $0.05/min PAYG fits contact-centre unit economics cleanly. Cartesia's subscription + credits model is harder to forecast for high-concurrency workloads where call volume varies dramatically.
What languages does Pulse support?
Strong English coverage and decent multilingual. For broad multilingual scale at predictable cost, Minimax is a better fit. For English-only telephony at the lowest cost, Pulse is the most cost-efficient choice we offer.
Can I use Smallest AI for outbound sales calls?
Yes — paired with our outbound-sales agent. The combination of low latency and low cost makes outbound voice economically viable at scale, especially in markets where per-call margins are thin.
What's the difference between Pulse and Lightning?
Both target telephony. Lightning is optimised for higher-concurrency burst patterns — peak-time call centres where thousands of calls land in the same minute. For most workloads Pulse is the default.
Ready to use Smallest AI inside your agents?
Get started in minutes — no API key management required.