ByteCosts
Tools
Calculator

RAG Cost Calculator

Estimate embedding, vector storage, reranking, and generation costs for a document search or support-answering workflow.

RAG workload
12,000
900
12%
1,800
6
650
700
450
60

Each query is embedded before the vector search

Estimated RAG run-rate
$594
per month, plus $0.216 initial indexing
Generation
$591
5K input / query
Embeddings
$0.026
12% refreshed
Vector store
$0.050
16,616 chunks
Reranking
.70
54K queries
Line itemMonthly
Generation$591
Monthly embedding refresh$0.026
Query embedding$0.065
Vector storage / query infra$0.050
Reranking .70
Vector and reranker rows are configurable infrastructure assumptions. Token-priced rows come from the selected ByteCosts model records.