Overview

Configure AI language models for use in Chat and Prompt features. MCP Explorer supports OpenAI, Azure OpenAI, Anthropic, Google, and local models.

Supported Providers

  • OpenAI: ChatGPT models (gpt-4o, gpt-4, gpt-3.5-turbo, etc.)
  • Azure AI Foundry: Azure OpenAI Service with deployment names
  • Anthropic: Claude models
  • Google: Gemini and other Google AI models
  • Local: Self-hosted or local models

Adding a Model

Step 1: Navigate to AI Models

Click ⚙️ AI Models in the navigation menu.

📸 Screenshot:

ai-models-page-overview

Description: Show the AI Models page with the list of configured models and “Add Model” button

Step 2: Click Add Model

Click Add Model to open the configuration form.

Step 3: Fill in Details

Required fields:

  • Name: Friendly name (e.g., “GPT-4o Production”)
  • Provider Type: Select from dropdown
  • Model Name: Model identifier (e.g., gpt-4o, claude-3-5-sonnet-20241022)
  • API Key: Your provider API key

Optional fields:

  • Endpoint: Custom API endpoint (required for Azure, optional for others)
  • Deployment Name: Required for Azure AI Foundry
  • System Prompt: Default system message (defaults to “You are a helpful assistant.”)
  • Note: Helper text or description

📸 Screenshot:

ai-models-form-filled

Description: Show the model configuration form with all fields filled (use placeholder values for API keys)

Step 4: Save

Click Save to add the model to your list.

Provider-Specific Configuration

📸 Screenshot:

ai-models-providers

Description: Show a blank form with default model OpenAI selected

OpenAI 

  Name: GPT-4o
Provider: OpenAI
Model Name: gpt-4o
API Key: sk-...
Endpoint: (leave blank for default)

Azure AI Foundry 

  Name: Azure GPT-4
Provider: Azure AI Foundry
Model Name: gpt-4
Deployment Name: my-gpt4-deployment
API Key: <azure-key>
Endpoint: https://<resource>.openai.azure.com

Anthropic 

  Name: Claude Sonnet
Provider: Anthropic
Model Name: claude-3-5-sonnet-20241022
API Key: sk-ant-...

Local 

  Name: Local LLM
Provider: Local
Model Name: llama-3-70b
Endpoint: http://localhost:11434
API Key: (optional)

Managing Models

Set Default Model

Select a model from the Default Model dropdown to set which model is used by default in Chat and Prompts.

📸 Screenshot:

ai-models-default-selection

Description: Show the default model dropdown with multiple models available

Edit a Model

  1. Find the model in the list
  2. Click Edit (pencil icon)
  3. Modify fields
  4. Click Save

Delete a Model

  1. Click Delete (trash icon)
  2. Confirm deletion

warning: Warning: Deleting a model removes its configuration. Any chats or prompts using this model will need a new model selected.

Using Models

In Chat

  1. Navigate to Chat tab
  2. Select model from dropdown at top
  3. Start chatting—model is used for all responses

In Prompts

  1. Get a prompt response
  2. Select model from LLM dropdown
  3. Click Execute with LLM
  4. Model processes the prompt

Troubleshooting

Authentication Errors

Solutions:

  • Verify API key is correct and active
  • Check API key has proper permissions
  • Ensure key hasn’t expired
  • Test key with provider’s playground/test tools

Model Not Available

Solutions:

  • Verify model name is spelled correctly
  • Check if model exists for your account/region
  • Review provider documentation for model names
  • For Azure: Verify deployment name matches

Slow Responses

Solutions:

  • Check network connection
  • Verify provider service status
  • Try a different model
  • Reduce conversation context length

Next Steps