> ## Documentation Index
> Fetch the complete documentation index at: https://fliqr.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Connect an AI Model Provider to Your Fliqr AI Account

> Connect OpenAI, Google Gemini, Anthropic Claude, or DeepSeek to power your AI Agents. Bring your own API key or use Fliqr AI's included quota.

Every AI Agent in Fliqr AI needs a language model to generate responses. You can use the included OpenAI quota that comes with your plan, or connect your own API key to use a different provider or model. Bringing your own key gives you full control over costs, model selection, and rate limits.

## Overview

Fliqr AI supports major AI providers — OpenAI, Google (Gemini), Anthropic (Claude), DeepSeek, and xAI. Each provider offers distinct trade-offs between cost, speed, and reasoning depth. You configure providers once in Settings; all AI Agents in your account can then reference them.

Your provider selection determines which models are available when you create or edit an AI Agent. You can use different providers for different agents, or set a single provider as your account default.

## Prerequisites

* An active Fliqr AI account on any paid plan
* An API key from your chosen provider (see provider links below)
* [AI Agents overview](/docs/core-concepts/ai-agents) — understand what an Agent is before configuring its model

## Supported Providers and Models

| Provider      | Models / notes                                                            |
| ------------- | ------------------------------------------------------------------------- |
| **OpenAI**    | GPT-4o, GPT-4o mini, o3-mini (plus speech/vision/image tools in Flows)    |
| **Google**    | Gemini 1.5 Flash, Gemini 1.5 Pro, Gemini 2.0 Flash (vision + image tools) |
| **Anthropic** | Claude 3 Haiku, Claude 3 Sonnet (vision)                                  |
| **DeepSeek**  | DeepSeek Chat                                                             |
| **xAI**       | Image generation and chat models available when connected                 |

<Note>
  The Pro plan includes 5 million tokens of GPT-4o mini per month at no additional cost. Once that quota is exhausted, Fliqr AI pauses AI Agent responses until the next billing cycle or until you connect your own OpenAI API key.
</Note>

## Connect a Provider

<Tabs>
  <Tab title="OpenAI">
    <Steps>
      <Step title="Open Integrations">
        Go to **Settings → Integrations → OpenAI**.
      </Step>

      <Step title="Enter your API key">
        Paste your OpenAI Secret Key into the **API Key** field. You can generate one at [platform.openai.com/api-keys](https://platform.openai.com/api-keys).
      </Step>

      <Step title="Select a default model">
        Choose the model that AI Agents will use when no per-agent override is set. For most deployments, start with **GPT-4o mini**.
      </Step>

      <Step title="Save">
        Click **Save**. Fliqr AI validates the key against the OpenAI API before confirming.
      </Step>
    </Steps>
  </Tab>

  <Tab title="Google Gemini">
    <Steps>
      <Step title="Open Integrations">
        Go to **Settings → Integrations → Google AI**.
      </Step>

      <Step title="Enter your API key">
        Paste your Google AI API key. You can generate one at [aistudio.google.com/app/apikey](https://aistudio.google.com/app/apikey).
      </Step>

      <Step title="Save">
        Click **Save**. Gemini models will now appear in the model selector when you create or edit an AI Agent.
      </Step>
    </Steps>
  </Tab>

  <Tab title="Anthropic Claude">
    <Steps>
      <Step title="Open Integrations">
        Go to **Settings → Integrations → Anthropic**.
      </Step>

      <Step title="Enter your API key">
        Paste your Anthropic API key. You can generate one at [console.anthropic.com/settings/keys](https://console.anthropic.com/settings/keys).
      </Step>

      <Step title="Save">
        Click **Save**. Claude 3 Haiku and Claude 3 Sonnet will appear as model options in your AI Agents.
      </Step>
    </Steps>
  </Tab>

  <Tab title="DeepSeek">
    <Steps>
      <Step title="Open Integrations">
        Go to **Settings → Integrations → DeepSeek**.
      </Step>

      <Step title="Enter your API key">
        Paste your DeepSeek API key from [platform.deepseek.com](https://platform.deepseek.com).
      </Step>

      <Step title="Save">
        Click **Save**.
      </Step>
    </Steps>
  </Tab>

  <Tab title="xAI">
    <Steps>
      <Step title="Open Integrations">
        Go to **Settings → Integrations → xAI** (label may appear as Grok / xAI).
      </Step>

      <Step title="Enter your API key">
        Paste your xAI API key from your xAI console.
      </Step>

      <Step title="Save">
        Click **Save**. Use xAI actions in Flows for supported tools such as [image generation](/docs/core-concepts/ai-agents/image-generation).
      </Step>
    </Steps>
  </Tab>
</Tabs>

## Choosing the Right Model

Different tasks call for different models. The table below summarizes the practical trade-offs:

| Goal                                 | Recommended model               |
| ------------------------------------ | ------------------------------- |
| High-volume customer support         | GPT-4o mini or Gemini 1.5 Flash |
| Complex reasoning / multi-step tasks | GPT-4o or Gemini 1.5 Pro        |
| Low latency responses                | Gemini 2.0 Flash                |
| Nuanced, long-form writing           | Claude 3 Sonnet                 |
| Cost-sensitive production            | GPT-4o mini or Gemini 1.5 Flash |

<Tip>
  For cost-sensitive deployments, GPT-4o mini and Gemini 1.5 Flash offer the best quality-to-cost ratio among current models. Start with one of these and upgrade only if you encounter reasoning gaps in your specific use case.
</Tip>

<Warning>
  API keys are stored encrypted, but they are visible to anyone with access to your Settings page. Do not share Settings access with team members you would not trust with your provider billing account.
</Warning>

## What's next

<CardGroup cols={2}>
  <Card title="Create an Agent" icon="robot" href="/docs/core-concepts/ai-agents/create-agent">
    Create the agent and attach business data.
  </Card>

  <Card title="Prompt engineering" icon="pen-to-square" href="/docs/core-concepts/ai-agents/prompt-engineering">
    Write system prompts that define behavior, tone, and constraints.
  </Card>

  <Card title="Knowledge sources" icon="book-open" href="/docs/core-concepts/ai-agents/knowledge-sources">
    Ground your agent in your own content.
  </Card>

  <Card title="AI Agents overview" icon="grid-2" href="/docs/core-concepts/ai-agents">
    Browse the full AI Agents section under Core Concepts.
  </Card>
</CardGroup>