SIMCOAI Lite
Cheapest per answer, for high-volume routine work.
Handles the ordinary questions - opening hours, where an order is, taking a message - at the lowest cost per answer.
Every SIMCOAI model is a configuration SIMCOAI builds and runs: its own instructions, its own context routing, its own deadlines, its own guardrails. That is what you choose between, and it is the part we control - so the engine underneath can change without changing what you picked or how it behaves.
Which one answers depends on your plan and on what is being asked. On a live call speed wins, because a caller left in silence hangs up; on a policy question depth wins.
Cheapest per answer, for high-volume routine work.
Handles the ordinary questions - opening hours, where an order is, taking a message - at the lowest cost per answer.
Fastest replies, tuned for live phone calls.
Answers in about a second so a caller is never left in silence. This is what answers your phone by default.
The everyday balance of speed and depth.
The default for chat, the dashboard assistant and anything that is not on a live call.
Deeper reasoning for harder questions.
Takes longer and thinks harder. Used where the answer matters more than a second of latency - policy questions, complex refunds, setup advice.
Speech straight to speech, with no transcription step in between.
An experimental mode where the model hears and speaks directly, rather than reading a transcript. Lower latency, and it keeps more of how something was said.
The voice your callers hear. Faster voices start speaking sooner; richer voices sound less like a machine. Your plan decides which are available, and you choose.
The quickest to start speaking. Used where the very first word matters more than anything else.
The everyday speaking voice, and what answers your phone by default. Fast enough that a caller is never left waiting.
A richer, more expressive voice for businesses where the greeting is the first impression.
The most expressive voice available, with the widest emotional range. The slowest to start, and the one that sounds least like a machine.
How SIMCOAI understands what a caller said - including when they have finished saying it, which is what stops the assistant talking over somebody.
Follows the rhythm of a conversation so the assistant knows when a caller has finished speaking, rather than waiting for a pause.
Highest-accuracy transcription, used where getting a name, an address or a reference exactly right matters most.
The general-purpose ear, used for recorded messages and anything not on a live call.
SIMCOAI builds on frontier models from established providers and adds its own instructions, routing, limits and safeguards on top. The names above describe what SIMCOAI configured and runs, not a claim to have trained a model from scratch.
Every provider SIMCOAI relies on is named in full in our privacy policy and GDPR and DPA page, because naming who processes your data is a legal requirement rather than a choice.
Models improve, prices move and providers have outages. Naming the configuration rather than the engine is what lets us move to a better one without you having to re-choose anything or re-learn what you bought.
Where a change would alter how your service behaves, it goes on our updates page, and anything affecting availability goes on the status page.
Start a free trial and put a real call through it. No card needed to try it.