Skip to content
All models
LiveWebSocket

Omni Agents

Managed phone agents with evals, telephony, and Recap.

A managed Omni offering for teams that want agents live fast: no-code setup, telephony packaging, evals, Recap, monitoring, knowledge bases, tools, and Trace-ready review loops on top of the realtime Omni engine.

Price
$0.08/min
Endpoint
Console + wss://api.pyai.com/v1/omni
Scope
omni:session
Model id
pyai-omni-realtime
Managed telephony$0.01/min

What you get

Launch the agent, not the stack

Describe the job, connect docs and tools, point a number at it, and run production calls without stitching vendors together.

Evals before volume

Test the conversations that matter before rollout, then monitor real calls as the agent scales.

Recap included

Every managed Agent call includes summaries, dispositions, next steps, and searchable call intelligence without a separate Recap line item.

Telephony packaged in

Bring your carrier or use managed telephony at a simple $0.01/min line item alongside the agent runtime.

Managed runtime Evals + monitoring Recap included Telephony packaging Knowledge bases + tools Trace-ready

Built on Omni's realtime engine. Managed runtime is $0.08/min; managed telephony is $0.01/min when used.

Start in minutes

Node
// Managed Agents use the same Omni realtime engine under the hood.
const ws = new WebSocket(
  "wss://api.pyai.com/v1/omni?agent_id=front_desk&format=pcm16&rate=24000",
  ["pyai-key." + apiKey],
);

FAQ

How is Agents different from Omni API?

Omni API is the developer realtime engine. Omni Agents packages that engine with no-code setup, evals, monitoring, tools, knowledge bases, and telephony help.

What does it cost?

$0.08/min for the managed agent runtime. If you use managed telephony, add $0.01/min for connected phone minutes.

Build with Omni Agents today.

Start free with $50.00 in credit - no card. Your test key works instantly.

No credit card - OpenAI-compatible - cancel anytime