Per-Query Billing for Search and RAG Pipelines
Bill every web search, vector retrieval, and document query your AI agents make. Per-query metering for Brave, Exa, Pinecone, Weaviate, and any search API.
How it works
import { settlegrid } from '@settlegrid/mcp'
const sg = settlegrid.init({
toolSlug: 'my-search-api',
pricing: {
defaultCostCents: 2,
methods: {
'web_search': { costCents: 2 },
'vector_query': { costCents: 5 },
},
},
})
const billedSearch = sg.wrap(async (args: { query: string }) => {
const results = await braveSearch(args.query)
return { content: [{ type: 'text', text: JSON.stringify(results) }] }
})Supported providers
SettleGrid works with any provider. Here are the most common ones for search & rag.
| Provider | Pricing |
|---|---|
| Brave Search | Web search API -- $0.002/query |
| Exa | Neural search -- $0.001-0.004/query |
| Tavily | AI-optimized search -- $0.003/query |
| Pinecone | Vector database -- per-read pricing |
| Weaviate | Vector search -- per-query pricing |
| Qdrant | Open-source vector DB -- self-hosted or cloud |
Why per-query billing?
Search and RAG pipelines naturally bill per-query because each retrieval has a discrete cost. Whether your agent calls a web search API, runs a vector similarity query, or fetches documents from a knowledge base, SettleGrid meters each operation independently. You can set different rates for different retrieval methods.
$4.3B
Total Addressable Market
6
Supported Providers
2 min
Setup Time
Frequently asked questions
Can I bill differently for web search vs vector queries?
How does billing work for multi-step RAG pipelines?
What about hybrid search (keyword + vector)?
Can I pass search costs through to my users?
Start billing search & rag today
Add per-query billing to your search & rag service in under 2 minutes. No upfront costs, no contracts.