> Text embeddings
// Converts text into vector embeddings for semantic search and similarity
Converts text into vector embeddings for semantic search and similarity. Agents can request it by capability ID ai.embeddings, or by plain intent — EntRoute resolves phrases like “embeddings”, “text embeddings” and “vectorize text” to this capability.
EntRoute currently lists 1 endpoint from 1 provider for ai.embeddings, priced at $0.001 per call, settled in USDC on Base. The quickest responds in about 29ms at the 95th percentile. Every endpoint below is re-probed every 10 minutes for a valid 402 payment challenge; the ordering blends measured success rate, latency and price rather than provider claims.
These phrases resolve to this capability with intent-based discovery.
curl -X POST https://api.entroute.com/discover \ -H "Content-Type: application/json" \ -d '{"capability_id": "ai.embeddings"}'
Use this capability in your agent or application:
Frequently asked questions about Text embeddings
How much does a text embeddings API call cost?
Endpoints listed for ai.embeddings on EntRoute are priced at $0.001 per call. Payment is per request — there is no subscription, account or API key. Prices shown are read from each endpoint's live 402 response, not from a provider-supplied catalogue.
How do AI agents pay for text embeddings endpoints?
Via the x402 protocol. The agent calls the endpoint, receives an HTTP 402 Payment Required response describing the price and recipient, signs a stablecoin payment (USDC on Base), and retries the request with proof of payment. The whole exchange takes seconds and needs no human approval.
How many providers offer text embeddings?
1 endpoint is currently listed for ai.embeddings across 1 provider. That number moves as endpoints are added and as failing ones are demoted.
Are these text embeddings endpoints verified?
Each one is probed every 10 minutes to confirm it still returns a well-formed 402 challenge on a payable network. Endpoints whose 7-day success rate drops below 50% are moved out of the ranked list into fallbacks with an explicit exclusion reason, so an agent can decide whether to retry them.