One voice model for phone agents. Not seven boxes stitched together.
Stop gluing STT, an LLM, turn detection, and TTS together like Vapi or Retell. Omni is one speech-to-speech model, built for telephony - audio in, audio out, $0.05/min all-in and OpenAI-compatible. Start free with $50 in credits. No card.
$50 in free credits - no card
Tap a voice - no signup, no key.
Want the full thing - Hear, Omni, code export?Open the Playground
One speech-to-speech model, built for the phone.
Vapi, Retell, and Bland stitch four vendors together - every hop adds latency and a markup. We own the whole stack on one engine, so turn-taking stays tight and the all-in minute lands at $0.05 as a result, not a discount.
Every layer is a separate vendor with its own markup, plus the latency of each hop. Stitched together, the all-in cost lands around $0.15-0.40/min.
Public all-in ranges, Jun 2026 - verify on each provider’s site.
We own the whole stack, so we sell at our cost plus margin— not four vendors’ prices plus margin. A reseller can only match $0.05 by losing money on every call. They can’t follow us down.
Live latency receipts, refreshing every couple of seconds. Verify these numbers against your own call scripts.
Powering voice & messaging for teams like
Plugs into the stack you already run
Developers ship in two lines. Teams launch with no code.
Whether you write the integration or describe the agent, PyAI meets you where you are.
For developers
Keep your OpenAI SDK. Change two lines, pass a PyAI key, and you have transcription, speech, and realtime agents - typed SDKs, idempotency, and stable error codes included. Nothing to provision: no agent registry, no deploy step - paste a key, open a socket, and you’re live.
curl https://api.pyai.com/v1/audio/speech \
-H "Authorization: Bearer $PYAI_KEY" \
-d '{"model":"pyai-voice","input":"Hello from PyAI."}' \
--output hello.mp3For teams
Describe a phone agent in plain English, ground it in your docs, give it tools - and point your number at it. A working receptionist that books, answers, and transfers, in minutes.
- No-code agent builder in the console
- Answers from your knowledge base
- Books, sends, and warm-transfers via tools
- Pennies per minute, everything included
Four products to build on. One free way to start.
Omni API, Agents, Trace, and Cast carry production voice work. Hear is free to start with, and Speak, Telephony, and Recap are the building blocks underneath.
Omni
The all-in-one AI voice agent model: hybrid speech-to-speech, fused LLM brain, emotion-aware voices, and tool calling.
Agents
Manage, evaluate & ship AI voice agents — no code, no engineering.
Cast
Emotional long-form TTS for podcasts, narration, and audiobooks.
Trace
Compliance & QA on every call, automatically.
Start free with Hear
Every account gets 10,000 transcription minutes a month, free. Ship with Hear, then scale into Omni, Agents, Trace, and Cast when you go to production - with Speak, Telephony, and Recap available as building blocks whenever you need them.
Watch a call flow through PyAI, end to end
Tap play. One front-desk call - ring to CRM - with every step powered by a PyAI model you can click into.
Acme Dental’s front-desk agent - every step powered by PyAI.
Hear <-> Speak repeats every turn, with barge-in - then the call becomes data.
Each step is a PyAI model. Press play to watch them hand off - or tap any stage to jump.
- Hear
pyai-hear - Omni brain
pyai-omni-realtime - Speak
pyai-voice
Transcript, summary, intent, and disposition - emitted as JSON when the call ends.
Built for the phone. Fast turn-taking. Drop-in compatible.
Telephony-grade voice done right - so your agents sound human, respond instantly, and ship without a rewrite.
Telephony-native
Hear is tuned for 8 kHz call audio and Omni is built for phone turn-taking and barge-in - not repurposed studio models.
Realtime and fast
Speak streams from the first byte (~32-98 ms on the warm path) and Omni answers with ~390 ms median turn-taking.
OpenAI-compatible
Keep your SDK and your error handling. Change base_url and key; everything else just works.
Grounded and tool-using
Bind knowledge bases and webhook tools so agents answer from your content and take real actions.
Honest billing
Per-second metering rounded once at the invoice, a clear rate card, and free credit to start. No surprises.
Predictable economics
Because it's one owned engine, the all-in minute is simple: Omni $0.05/min and transcription $0.003/min, per-second AI billing with telephony split out separately.
SOC 2 compliant
Independently audited security controls protect your data - and your callers'.
HIPAA-ready
A Business Associate Agreement and custom data terms for regulated teams.
Tamper-evident audit trail
Trace stamps every call with a verifiable audit hash and findings that cite the rule.
99.99% uptime
Real-time status and incident history at status.pyai.com.
OpenAI-compatible
Point your existing SDK at our base URL - keep your code and error handling.
A human answers
Reply to any email and a person reads it - not a ticket queue.
Teams book more - and resolve faster - with PyAI.
From AI appointment booking to compliance on every call, here's what customers see.
“Our AI Agents book appointments 24/7 - that drove 57% incremental revenue for our customers. It's our hottest-selling product.”
“We run healthcare and TCPA-heavy lines. Trace auto-scores every call against our rule packs, so compliance stopped being a fire drill.”
“Time to first reply is under a minute and CSAT is at an all-time high - while our cost per resolution keeps dropping.”
See where your voice AI budget leaks
Compare labor, platform fees, and all-in agent minutes side by side. The two best-value options are PyAI.
One headline price, then clear packages.
Start free with $50 in credits and 10,000 free speech-to-text minutes every month. Pay for Omni, Agents, Trace, and Cast as you use them.
Realtime voice agent API
Manage, evaluate & ship agents — no code
Compliance, QA, evals
Async emotional TTS
Hear gives developers 10,000 free transcription minutes every month. Telephony is $0.01/min; Recap is a $0.03/min add-on for Omni API, Hear, and Agents; KB Context is $0.003/min.
in free credits to start. Plus 10,000 free speech-to-text minutes every month and $5.00/month recurring credit. No card.
Claim your free creditsUsing Cursor, Claude Code, Codex, or Lovable?
Start with one prompt. We hand your agent the API contract, the SDKs, and a free test key - so it writes correct PyAI code on the first try.
Questions, answered
Is PyAI OpenAI-compatible?
Yes. Point your existing OpenAI SDK at https://api.pyai.com/v1 and pass your PyAI key. Transcription and speech use the same request and response shapes, and errors come back in the OpenAI envelope.
How much does it cost?
Hear (transcription) gives you 10,000 minutes free every month; beyond that it's $0.003/min. Omni API is $0.05/min all-in, and the Agents feature (manage, evaluate & ship agents, no code) is $0.08/min. You also start with $50.00 in free credits plus $5.00/month - no card.
Is it good for telephony?
That's the focus. Hear is tuned for 8 kHz call audio, and Omni does end-to-end phone agents with ~390 ms median turn-taking and barge-in.
Do I need a credit card to start?
No. Sign up and your sandbox test key works instantly with free credit. Add funding only when you're ready to send live traffic.
Ship phone agents on one speech-to-speech model.
Start with a free key, 10,000 Hear minutes every month, and $50 in credits. Build on the same OpenAI-compatible stack you take to production with Omni, Agents, and Trace - natural turn-taking, telephony-native, live in an afternoon.
No credit card - OpenAI-compatible - cancel anytime
