Embedding/Cache
Embedding Cache
Reduce duplicate cloud inference requests.
Enabled
Hit Rate
62.4%
last 24h
Entries
842K
Storage Size
3.4 GB
Estimated Savings
$18.20
last 24h
Hit / Miss Over Time
00:0006:0012:0018:00Now
Configuration
- Enabled
- Yes
- Retention
- 30 days
- Max entries
- 2,000,000
- Max storage
- 10 GB
- Cleanup schedule
- Daily 03:00 UTC
- Evictions (24h)
- 1,248
Clearing embedding cache removes cached inference results only. Embeddings stored by Memory Service are not affected.