Output nodes allow you to pass data out of a pipeline.
Node Inputs
Output: the data you are looking to output from the pipeline. The available types for this field are Text, Streamed Text, File, Image, Audio, and JSON depending on what you choose for the Type parameter (see right below).Node Parameters
Type: The data type that the output node will output. Use the Type dropdown to signify the data type. The available types are: Text (default), Streamed Text, File, Image, Audio, and JSON.Streamed Text will display any text as soon as it is generated by an LLM as opposed to waiting for the entire output to finish generating before showing the output.
To turn on streamed text, toggle to True the parameter “Stream Response” in the gear of any LLM node.
Format Output (Default True): only applicable to Text and Streamed Text. If toggled to True, it will format the text in Markdown.
Node Outputs
The output node does not have any node outputs.Considerations
In addition to passing data out of a pipeline, a common use case for output node is to debug workflows by connecting the Output node to intermediary nodes. ThisExample
The below example is a pipeline for chatting with a knowledge base. The LLM node is connected to the output node so the generated text can be shown to the user.- Input Node: Represents the user message
- Knowledge Base Reader Node: Queries the knowledge base semantically
- Search Query:
{{input_0.text}}
- Search Query:
- LLM Node: Responds to the user question
- System (Instructions):
Answer the question based on the context. - Prompt:
Question: {{input_0.text}} Context: {{knowledge_base_1.chunks}}
- System (Instructions):
- Output Node: Displays the response
- Output:
{{openai_0.response}}
- Output:
