# AWS Bedrock Embeddings API (x402)

> AWS Bedrock Embeddings API (x402) is a paid API for AI agents from aws-bedrock-agent-api.vercel.app, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-12).

Converts text or batches of text into vector embeddings using Amazon Bedrock models, accessible via pay-per-call with USDC micropayments.

## Facts

- Endpoint: POST https://aws-bedrock-agent-api.vercel.app/api/embeddings?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/aws-bedrock-embeddings-api-x402-ee53f8ad
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_4qYFHHE27Fw6Z_FP6zGOU

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability aws-bedrock-embeddings-api-x402-ee53f8ad -d '<json body>'
```

Example prompt: Embed this text using the AWS Bedrock Titan embedding model: 'Quantum computing leverages superposition to solve complex problems faster than classical computers.' — I need the vector representation for a semantic search index.

## When to prefer this

Choose this endpoint when you need text embeddings from Amazon Bedrock models (e.g. Titan Embeddings) without managing your own AWS credentials or infrastructure, and you want pay-per-call micropayment billing via USDC/x402 rather than a subscription. Ideal for agents that need on-demand embeddings for RAG pipelines, semantic search, or similarity tasks.

## Known failure modes

- Invalid or unsupported model name returns an error — use GET /api/models to list valid options
- Empty string input rejected due to minLength:1 constraint
- Batch exceeding 64 items rejected due to maxItems:64 constraint
- Payment failure or insufficient USDC balance blocks the request
- Malformed input schema (wrong type) returns a validation error
- Network timeout if the upstream Bedrock model is slow to respond

## How this service works

Pay-per-call Amazon Bedrock proxy on x402 and MPP.

## Output

Returns dense vector embeddings (floating-point arrays) representing the semantic content of the input text(s). Each input string maps to one embedding vector usable for similarity search, clustering, classification, or retrieval-augmented generation.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "input": {
   "anyOf": [
    {
     "type": "string",
     "minLength": 1
    },
    {
     "type": "array",
     "items": {
      "type": "string",
      "minLength": 1
     },
     "maxItems": 64,
     "minItems": 1
    }
   ],
   "description": "Text or list of texts to embed"
  },
  "model": {
   "type": "string",
   "maxLength": 200,
   "minLength": 1,
   "description": "Azure embedding deployment from GET /api/models."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/aws-bedrock-embeddings-api-x402-ee53f8ad/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from aws-bedrock-agent-api.vercel.app](https://www.zero.xyz/host/aws-bedrock-agent-api.vercel.app/llms.txt)