5 cookbooks covering every aspect of conversation memory with Valkey - from basic chat history to semantic search, caching, and agent state management.
Connect with GLIDE, store chat messages in a LIST, retrieve conversation history, and auto-expire sessions with TTL.
Track session metadata with HASH - user ID, model, token count, timestamps. List active sessions and implement sliding window history.
Store conversation embeddings with JSON.SET and search by meaning with FT.SEARCH KNN. Find relevant past context across all sessions.
Cache LLM responses by meaning, not exact match. Skip expensive API calls when a similar question was already answered.
Checkpoint agent reasoning with HASH, log tool calls to STREAM, and resume multi-step workflows after interruption.
Watch Valkey commands fire in real-time as conversations are stored, searched, and cached.
Open Interactive Demo