← What does your AI stack cost?

How much does a RAG app cost to run?

A retrieval-augmented app touches four paid layers. Here's the cheapest viable combination for a reference workload, with a note on each.

50,000 requests / month · 4000/500 in/out · 20,000 docs · 600 tok/doc · 55% cached
LLM API
gpt-oss-20bvia Darkbloom
$6.50 /mo

The generation call. Input dominates (retrieved context), so cached-input pricing and a cheap-but-capable model matter most.

next: Qwen3.7 Flash $6.61compare all ↗

or self-host: 1× A40 at ~55% GPU use $358/mo· pays off above ~2.8M requests/mo

Embeddings
text-embedding-3-smallvia OpenAI
$0.08 /mo

One vector per chunk at index time, plus a small embedding per query. The initial index is a one-off; ongoing cost is tiny.

next: voyage-3.5-lite $0.08compare all ↗
Vector DB
Zilliz Cloud (Serverless)via Zilliz
$0.12 /mo

Stores and searches the vectors. Serverless bills storage + queries; dedicated bills compute time - the crossover is around a few million vectors.

next: turbopuffer $0.12compare all ↗
Object storage
B2via Backblaze
$<0.01 /mo

The raw documents sit in object storage. Cheap; the swing factor is egress if you re-index often.

next: Object Storage $<0.01compare all ↗
Observability
Arize Phoenix
$0 /mo

Tracing and eval. Langfuse and Helicone have generous free tiers - $0 until real volume.

next: Helicone $0compare all ↗
Total, your workload≈ $6.70 /mo
Tune it for your numbers ↗

Live prices, updated 2026-09-15. A model, not a quote - each layer is priced at its cheapest provider; embeddings amortise the initial index over 12 months. Confirm on each provider's site.

Concepts on this page