$

SemCache

A semantic caching layer that cuts LLM API spend.

text-embedding-3-small+gpt-4o-mini

Try a group top to bottom. The first question fills the cache; the rephrasings underneath should hit it.

Refunds

Hours

Shipping

Savings dashboard

live totals across every query this cache has served

Total requests

2

Cache hit rate

50%

LLM calls avoided

1

Total saved

$0.000042

Time saved

2.2 s

Cache contents

similarity threshold 0.44
Stored queryHitsOriginal cost

What is your refund policy?

gpt-4o-mini

1$0.000042