Reduce LLM API Costs with a Redis Semantic Cache for Repeated Questions

Cache LLM answers in Redis with Kestra to cut OpenAI costs. Normalizes questions into keys and serves instant responses on cache hits.