# Nova AI Inference — Structured Data Extraction

> Nova AI Inference — Structured Data Extraction is a paid API for AI agents from novva.orbonomy.xyz, paid per call via x402, $0.1/call, status unknown (last checked 2026-09-15).

Extracts structured data from text using AI inference, returning typed fields with a confidence score, paid per-call in USDC on Base.

## Facts

- Endpoint: POST https://novva.orbonomy.xyz/api/extract
- Price: $0.1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/nova-ai-inference-structured-data-extraction-359e3323
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_7cigVfWogbh-Xi8Pk3tKj

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 nova-ai-inference-structured-data-extraction-359e3323 -d '<json body>'
```

Example prompt: Extract the vendor name, invoice date, total amount, and line items from this invoice text and return them as structured fields: 'Invoice from Acme Corp, dated March 15 2025, total $4,320.00, for 3 units of Widget A at $1,440 each.'

## When to prefer this

Use this endpoint when you need to turn free-form or semi-structured text into a typed JSON object with a defined schema and want a confidence score alongside the result. It is especially suited for invoice parsing, entity extraction, form filling from prose, and any task where the output shape is known in advance. Prefer it over general chat endpoints when the goal is structured extraction rather than conversational generation.

## Known failure modes

- Missing required 'text' field returns a validation error
- Schema mismatch causes low-confidence or incomplete extraction
- Ambiguous or contradictory instructions reduce extraction accuracy
- Payment failure in USDC returns 402 status before processing begins
- Very long input text may exceed model context limits

## How this service works

Pay-per-call AI inference. Chat, reason, translate, analyze, generate. USDC on Base.

## Output

Returns a JSON object containing an 'extracted' field with typed values matching the requested schema, a confidence score (0-1), the model name used, and token usage stats, wrapped in a success flag and metadata envelope.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "text"
 ],
 "properties": {
  "text": {
   "type": "string",
   "description": "Text to extract data from"
  },
  "schema": {
   "type": "object",
   "description": "Expected output schema"
  },
  "instructions": {
   "type": "string",
   "description": "Additional extraction instructions"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "data": {
   "type": "object",
   "properties": {
    "model": {
     "type": "string"
    },
    "usage": {
     "type": "object"
    },
    "extracted": {
     "type": "object"
    },
    "confidence": {
     "type": "number"
    }
   }
  },
  "meta": {
   "type": "object"
  },
  "success": {
   "type": "boolean"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/nova-ai-inference-structured-data-extraction-359e3323/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from novva.orbonomy.xyz](https://www.zero.xyz/host/novva.orbonomy.xyz/llms.txt)