Streaming RAG answers for real-time response generation with document citations and progressive content delivery.
Key Features:
Note: Example requests/responses shown here are illustrative. When using the live “Try it” tool, you may see two responses: an HTTP status preview from the client and the API’s streaming output.
Your API key from account settings
Your bot ID from account settings
The question to answer using indexed documents
"ما هو الاسم الكامل لمحمد صلاح؟"
Names of document indexes to search
["Mohamed Salah"]Number of document chunks to retrieve
1 <= x <= 5010
Maximum number of documents to process
1 <= x <= 5010
Controls creativity in answer generation
0 <= x <= 10.1
Maximum tokens in generated answer
1 <= x <= 40961024
Whether to update global context
false
Whether to stream the response Must be true for streaming responses
true
Streaming RAG response with progressive content and citations
Server-sent events with answer chunks and metadata