The OpenAI connector brings the power of Large Language Models (LLMs) directly into your automation pipelines. It is designed to go beyond simple chat interfaces, focusing on reliable, structured data extraction and decision making.
Key Capabilities
Reliable JSON: We automatically handle retry logic and validation to ensure that the LLM outputs valid JSON that your downstream tools can consume.
Agentic Workflows: By combining Function Calling with our workflow engine, you can build autonomous agents that can read data, make decisions, and take actions.
RAG Ready: Built-in support for generating embeddings makes it easy to implement Retrieval-Augmented Generation (RAG) architectures.
Features
Structured Outputs
Force JSON output from LLMs to ensure downstream compatibility.
Function Calling
Expose your workflow steps as functions to the LLM.
Context Management
Automatically manage token windows and conversation history.
Embeddings
Generate vector embeddings for semantic search and RAG.
Ready for automation that just works?
Tell us about your process. We will tell you if and how dataflows can help.