Real-time messaging for AI workloads. Broadcast LLM tokens, fan out agent events, and build durable task queues - all with Valkey Pub/Sub and Streams.
6 guides from basic PUBLISH/SUBSCRIBE to production consumer groups
PUBLISH, SUBSCRIBE, and pattern matching
Broadcast token-by-token LLM responses to multiple clients
XADD, XREAD, XRANGE - durable message queues
Distributed processing with XREADGROUP and acknowledgments
One producer → many consumers for AI events and notifications
Backpressure, monitoring, reconnection, dead letter queues
Publish messages, subscribe to channels, stream LLM tokens, and test fan-out patterns
All cookbooks, demo, and example code for Pub/Sub & Streaming patterns.