Tool setup

Open WebUI

Open WebUI is a self-hosted chat front end that talks to any OpenAI-compatible endpoint: add AdvancedMind AI as a connection with a base URL and key.

Before you start

You need a running Open WebUI instance, an active API key, and credits on the org. Create a key under API keys and copy the secret; it is shown once. A standard-scoped key can use either Klara preview whenever that route is available.

1. Add the connection

In Open WebUI, open Admin Panel / Settings / Connections, find the OpenAI API section, and add a connection. Set the base URL to the gateway and paste your key. The base URL must end in /v1; Open WebUI appends /chat/completions and /models to it.

Connection settings

# Admin Panel -> Settings -> Connections -> OpenAI API -> add a connection
API Base URL:  https://api.advancedmind.ai/v1
API Key:       amai_sk_...          # the secret shown once at key creation
# Prefix / connection name (optional): AdvancedMind AI

To configure it at startup instead, set the same two values as environment variables when you launch the container or process. The in-UI connection takes precedence if both are present.

bash

# or configure the connection at startup with environment variables
OPENAI_API_BASE_URL="https://api.advancedmind.ai/v1"
OPENAI_API_KEY="amai_sk_..."

Use the connection's Verify button to confirm Open WebUI can reach the gateway. If it fails, check that the URL ends in /v1 and that the key is active and has not been revoked on the API keys page.

2. Select models

Open WebUI fetches the model list from /v1/models using your key. After saving the connection, the AdvancedMind models appear in the model picker at the top of a new chat. To limit what users see, hide models under Admin Panel / Settings / Models:

Klara Base Preview v2: harder reasoning, coding, research, and agent work with capability retention prioritized.
Klara Base Preview v1: a more direct default for interactive chat.

Per-model pricing is on the pricing & credits page. The Klara previews stream, so replies render token by token in the chat.

Usage and cost

Open WebUI sends the full conversation back as context on every turn, so input tokens and cost grow as a chat gets longer. Start a new chat when you switch topics to keep each request lean. Every request is metered against your org balance and recorded with its model, token counts, and final cost on the Usage page, regardless of which Open WebUI user sent it. Top up or set a low-credit alert on Billing. A request that would run past your balance is blocked before any provider call.

Shared instances and keys

A connection key is shared by everyone who uses that Open WebUI instance, and all of their requests draw on the same org balance. Give the instance its own key, separate from your personal or CI keys, so you can revoke it without disrupting anything else, and set a monthly spend limit on it to cap total spend. Use a research key only when the shared instance is intentionally operating under an approved Research Mode scope. Manage and revoke keys on the API keys page.

Next: streaming, error handling, or the Continue setup guide.