Enhance Power Platform with Mistral Large deployment for powerful AI integration. #Mistral #MistralAI #PowerPlatform

Integrating Mistral Large into Power Platform is like adding the secret ingredient to Power Automate and Power Apps, creating a full history context conversation. The mistra model offers natural language responses, providing a source of conversational insight and chatter for a more immersive experience.😎 #AIStudio #Azure #VendorMistralAI

In today’s demonstration, we will be covering the deployment of the language model Mistral Large into AI using AI studio Azure. This model, now available in Azure, will be integrated into the Power Automate and Power Apps for full history context conversation.

Setting Up Azure AI Hub Resource 🌐

Firstly, we need to set up an Azure AI Hub resource in the Azure portal. This requires setting up a new Resource Group, naming it ‘Mistral AI Hub resource’, and selecting the appropriate deployment location.

• Azure AI Hub Resource Setup
  | Feature | Description |
  |——————|———————————–|
  | Resource Group | Mistral AI Hub resource |
  | Deployment | East US / Central / France Central |
  | Name | Mistral Large Language Model |

Deployment of Mistral Language Model ⚙️

After creating the AI Hub resource, we proceed to deploying the Mistral language model into the Azure AI Studio. Despite challenges associated with deployment, we can overcome them by ensuring a default credit card is associated with the subscription.

• Creating AI Project and Deployment
  | Feature | Description |
  |——————|—————————————|
  | Project | Large LM Project |
  | Deployment | Instance of the Large Model |
  | Deployment Name | DEMO BIR 365 |

Integration with Power Automate 🤝

With the deployment complete, we move on to integrating the model with Power Automate by configuring a manually triggered flow. This includes setting up an HTTP action and custom connectors, adjusting URLs and authorizations, as well as specifying the message content and getting responses from the HTTP action.

• Power Automate Integration
  | Feature | Description |
  |——————|——————————————|
  | Input Setting | Manually triggered flow with user input |
  | URL Configuration| HTTP action customization for deployment |
  | Message Content | Modifying message prompts and user inputs|

Power App for Conversation History 💬

Finally, we explore the integration with Power Apps to showcase the history and context of the conversation. By demonstrating a simple app with text input and gallery display, we are able to map out question and answer pairs using collections from the flow.

• Power App Integration for Conversation History
  | Feature | Description |
  |——————|————————————————-|
  | Flow Calling | App triggers flow to call Mistral API History |
  | History Context | Collection of conversation pairs from the flow |
  | User Interface | Displaying the conversation history in the app |

In conclusion, the deployment and integration of the Mistral Large language model with Power Platform carries immense potential for creating conversational AI experiences. The seamless flow of information and the ability to build upon conversation history enriches the user engagement and overall user experience.

Key Takeaways 📝

• Deployment of Mistral Large into Azure AI Studio
• Configuration of Power Automate for message responses
• Integration of Power Apps for conversation history display

Hope you enjoyed this demonstration! For any further inquiries or request for code explanation, drop a comment. Stay tuned for more updates!