# Provision PHOENIQS Chat

This guide covers provisioning a dedicated PHOENIQS Chat instance for your company: its own URL, its own databases, and your own sign-on.

If your company already runs a chat instance and you only want to start using it, you do not need this page. Go to the Quick Start instead.


# Prerequisites

Before you begin, make sure you have:

  • An active Phoeniqs Console account with permission to make purchases
  • A company profile associated with your account

If you already have a fully provisioned Base LLM API in the resource group you want to use, skip to Step 2.


# Step 1: Purchase and provision a Base LLM API

PHOENIQS Chat runs on the Base LLM API. You need an active, fully provisioned Base LLM API in the same resource group before you can provision a chat instance. See Base LLM API for plans and pricing.

  1. Sign in at https://console.phoeniqs.com.
  2. Open Products from the left navigation, or click Add Subscription from the Dashboard.
  3. Select AI Model Service, then choose Base LLM API.
  4. Click Add to cart, then Proceed to Checkout.
  5. Complete checkout and confirm your order.

For the general checkout flow, see Buy a Subscription.

Then provision it:

  1. Navigate to Resource Groups in the left-hand menu and open the resource group that will hold your chat instance.
  2. Expand the Subscriptions list and find your Base LLM API subscription with status Ordered.
  3. Click Provision Service on that row.
  4. Enter a subscription alias, a readable label you will see in later dropdowns — for example MaaS-API-for-Finance-Team-Phoeniqs-Chat.
  5. Click Provision.

Provision service dialog for Base LLM API
Provision service dialog for Base LLM API

Provisioning takes a few seconds. Wait until the subscription shows as provisioned before moving on; an unfinished subscription does not appear in the chat provisioning dropdown.

You do not need to create an API key yourself. The platform creates one on this Base LLM API and attaches it when you provision the chat instance in Step 3. With one subscription you can create as many API keys as you want and re-create them at any time.


# Step 2: Purchase the PHOENIQS Chat subscription

  1. Open Products from the left navigation, or click Add Subscription from the Dashboard.
  2. Select PHOENIQS AI Platform, then choose your PHOENIQS Chat plan. See Subscriptions for what each plan includes.
  3. Click Add to cart, then Proceed to Checkout.
  4. Complete checkout and confirm your order.

For the general checkout flow, see Buy a Subscription.


# Step 3: Provision PHOENIQS Chat

  1. Navigate to Resource Groups in the left-hand menu and open the resource group holding your Base LLM API.
  2. Expand the Subscriptions list and find your PHOENIQS Chat subscription with status Ordered.
  3. Click Provision Service on that row.
  1. In the Provision service dialog:
    • Enter a subscription alias, a readable label you will use to identify this instance later.
    • Select a MaaS Instance — this is the Base LLM API from Step 1.
  2. Click Provision.

Provision service dialog for PHOENIQS Chat
Provision service dialog for PHOENIQS Chat

Provisioning normally takes 5 to 10 minutes. Watch the subscription status in the console until it shows as provisioned.


# Step 4: Open your instance

Once the subscription shows as provisioned, click Launch Product on the PHOENIQS Chat row.

Launch Product on a provisioned PHOENIQS Chat subscription
Launch Product on a provisioned PHOENIQS Chat subscription

Your instance opens at its own URL and you sign in with your PHOENIQS account. Paid plans use enterprise SSO, so you sign in through your company identity provider rather than typing a password into the chat.


# What happens next

  • Your chat instance draws inference from the credit pool of the Base LLM API you selected. All keys on that subscription share the same pool. See Base LLM API.
  • To add users, share your instance URL with them. Access is controlled through your identity provider.
  • To increase the credit allowance, see How to add LLM Credits.

# If provisioning fails

What you see What it means What to do
The MaaS Instance dropdown is empty You have no provisioned Base LLM API in this resource group, or you only have an Evaluation plan Purchase and provision a Base LLM API first, or switch resource group
Your Base LLM API is missing from the dropdown It has not finished provisioning, or it sits in a different resource group Wait for provisioning to finish, or open the correct resource group
"The selected MaaS key is invalid" The Base LLM API behind it is not responding, or the subscription is not healthy Check the status dashboard, then retry
Stuck in progress, then fails The instance could not be built Contact support with your subscription ID
Fails immediately The request was rejected The error message names the reason

# Removing a chat instance

A chat instance cannot be removed while pre-built assistants are still installed on it. Remove the assistants first, then the chat instance.

Removing a chat instance does not remove the API key it created on your Base LLM API. That key stays on the Base LLM API subscription and should be deleted separately if you no longer need it.

For the cancellation procedure, see How to cancel a subscription.


# See also

Quick Start — your first conversation
../quick-start/
PHOENIQS Chat subscription plans
../subscriptions/
Model Service subscription plans
../../../maas/subscriptions/