Summary Processor #
Generates AI-powered document summaries and insights with structured analysis.
Configuration #
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
message_field | string | documents | The field in the pipeline context containing the documents to process | |
output_queue | object | null | Optional queue configuration for sending processed documents to a output queue | |
model_provider | string | Yes | - | ID of the LLM provider |
model | string | Yes | - | Name of the LLM model |
model_context_length | uint32 | Yes | - | Minimum context length (min: 4000 tokens) |
min_input_document_length | uint32 | No | 100 | Minimum bytes to process a document |
max_input_document_length | uint32 | No | 100000 | Maximum document size to process |
ai_insights_max_length | uint32 | No | 500 | Target length for AI insights (in tokens) |
Example #
- document_summarization:
model_provider: openai
model: gpt-4o-mini
model_context_length: 4000
min_input_document_length: 100
max_input_document_length: 100000
ai_insights_max_length: 500