Multi-stream and Burst Mode

VSS supports multiple parallel requests for summarization, Q&A and alerts. This is supported for both files and live streams (RTSP). Use cases include analyzing multiple short video files (burst mode) or analyzing and monitoring multiple camera feeds (multi-stream).

Clients can call any of the APIs including Summarization (POST /summarize) and Q&A (POST /chat/completions) in parallel for different files / live streams from different threads or processes. Clients do not have to worry about queuing and synchronization since the VSS backend will take care of queuing and scheduling the requests. VSS with the help of CA-RAG will also be responsible for maintaining contexts for each of the sources.

By default, the sources are processed independently. There is no context sharing across the sources.

Following table describes the method for multi-stream processing for various modes of using VSS.

Mode	Method
Gradio UI	Open a new tab for each source (file or live-stream) and summarize, configure alerts and chat about the source. However, due to a browser limitation, loading multiple Gradio sessions in the same browser may cause Gradio sessions to get stuck or appear to be slow.
VSS Reference Python CLI / curl	Run VSS Reference Python CLI / curl commands for add / summarize & configure alerts / chat / delete in a new shell for each source
Programmatic API integration	REST APIs for add / summarize & configure alerts / chat / delete can be called programmatically from a different thread or process per source.