Model Integration
Vantage supports 5 AI providers with dozens of models across three performance tiers. You can choose your preferred provider and specific model, or let the platform automatically select the best model for each task.
Supported Providers & Models
Every provider supports three model tiers — Fast (quick responses, lower cost), Pro (balanced capability), and Deep Thinking (maximum reasoning power).
| Provider | Fast | Pro | Deep Thinking | Setup |
|---|---|---|---|---|
| Intuidy AI | GPT-4.1 Mini | GPT-4.1 | o3 | No setup required — available immediately |
| OpenAI | GPT-4.1 Mini | GPT-4.1 | o3 | API key required |
| Claude (Anthropic) | Claude Sonnet 4 | Claude Sonnet 4.6 | Claude Opus 4.6 | API key required |
| Gemini (Google) | Gemini 2.0 Flash Lite | Gemini 2.5 Flash | Gemini 2.5 Pro | API key required |
| Mistral | Mistral Small | Mistral Large | Magistral Medium | API key required |
Additional Supported Models
Each provider supports its full model family. The platform is compatible with any model matching these families:
| Provider | Supported Model Families |
|---|---|
| OpenAI | GPT-4o, GPT-4o Mini, GPT-4 Turbo, GPT-4.1, GPT-4.1 Mini, GPT-5, o1, o1 Mini, o3, o4 |
| Claude | Claude 3 Haiku, Claude 3.5 Sonnet, Claude Sonnet 4, Claude Sonnet 4.6, Claude Opus 4.6 |
| Gemini | Gemini 1.5 Flash, Gemini 1.5 Pro, Gemini 2.0 Flash, Gemini 2.0 Flash Lite, Gemini 2.5 Flash, Gemini 2.5 Pro, Gemini 3, Gemini Experimental |
| Mistral | Mistral Small, Mistral Medium, Mistral Large, Pixtral, Codestral, Devstral, Magistral, Ministral |
Org-level overrides: Administrators can map any supported model to any tier, so your organization can use a custom model configuration across the platform.
Configuring Your AI Provider
- Navigate to Settings → AI Features → Intuidy AI.
- Select your preferred AI Provider from the dropdown.
- If the provider requires an API key, navigate to the Integrations page and connect the provider.
- Select the specific AI Model you want to use.
- Verify the Connection Status shows as connected.
- Click "Save AI Settings".
Where AI Provider Is Used
Your selected AI provider powers all AI features across the platform:
| Feature | How the Provider Is Used |
|---|---|
| Global AI Assistant | Responds to your questions and generates queries |
| AI Agents | Powers deployed agents with tool-calling, data access, and conversation |
| Tile Summaries | Generates plain-language explanations of tile data |
| Tile AI Chat | Powers per-tile conversations about specific datasets |
| AI Workflow Nodes | Processes data through enrichment, formatting, compliance, etc. |
| AI Build Tile | Creates tile configurations from natural language descriptions |
| AI Auto-Fill | Scans company websites during onboarding |
Integration Setup
To use a third-party AI provider, you need to add your API key:
- Go to the Integrations page.
- Find the AI provider card (OpenAI, Claude, Gemini, etc.).
- Click "Connect".
- Enter your API key.
- Click "Save" to verify the connection.
Each provider card shows the connection status so you know the key is valid.
Choosing a Provider
| Factor | Recommendation |
|---|---|
| No setup needed | Use Intuidy AI (default) — production-ready immediately |
| Best general performance | OpenAI GPT-4.1 or Claude Sonnet 4.6 |
| Maximum reasoning | OpenAI o3, Claude Opus 4.6, or Gemini 2.5 Pro |
| Cost efficiency | Gemini 2.0 Flash Lite or Mistral Small for high-volume workloads |
| Data privacy | Mistral (European) or evaluate each provider's data policies |
| Multimodal (images/vision) | OpenAI GPT-4o, Gemini 2.0 Flash, or Pixtral |
| Code generation | Codestral or GPT-4.1 |
You can switch providers at any time — the change takes effect immediately for all new AI requests.