The June 2025 release of Application Integration includes the following new features and enhancements.
New recipes
Application Integration provides the following new recipes:
Databricks Mosaic AI Chat with File
Use the recipe to upload a file with context, provide a user prompt, read the text from the file, and answer the user's questions using the file content and the Databricks Mosaic AI LLM.
Generate Image with NVIDIA NIM
Use the recipe to create images that align with the text-based descriptions provided in natural language prompts.
Loan Processing
Use the recipe to evaluate a loan request and approve or reject it based on the applicant's credit score.
Patient Food Vendor Recommendation using Databricks Mosaic AI
Use the recipe to extract food allergy data from FHIR and find vendors with suitable food for the patient using Databricks vector index.
Simple RAG Consumption with Microsoft Teams AI Agent
Use the recipe to receive a user query from Microsoft Teams, search for relevant context in a vector database, send the query and context to a Large Language Model (LLM), and return the generated response back to the Microsoft Teams channel.
Simple RAG Consumption with Slack AI Agent
Use the recipe to receive a user query from Slack, search for relevant context in a vector database, send the query and context to a Large Language Model (LLM), and return the generated response back to the Slack channel.
For more information about these recipes, see the corresponding recipe guides.
New NVIDIA NIM Connector
Use NVIDIA NIM Connector to leverage the power of NVIDIA's accelerated computing to access and run a variety of AI models.
For more information, see the NVIDIA NIM Connector guide.