# AI Synthetic Training Examples Generator

> AI Synthetic Training Examples Generator is a paid API for AI agents from market.datapackvibe.com, paid per call via x402, $0.04/call, status unknown (last checked 2026-10-21).

Generates three short prompt/completion draft pairs from a task description and reference text using a local small language model, intended for training data bootstrapping.

## Facts

- Endpoint: POST https://market.datapackvibe.com/x402/ai-synthetic-examples?utm_source=zero.xyz
- Price: $0.04/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-21
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/ai-synthetic-training-examples-generator-3637bb77
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_-vJNZmx7fUotwGPFZqPby

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 ai-synthetic-training-examples-generator-3637bb77 -d '<json body>'
```

Example prompt: Can you generate three synthetic prompt/completion training examples for a customer refund Q&A task? The reference material is: 'Northwind support opens Monday to Friday at 09:00 and 17:00. Customers can request a refund within thirty days.'

## When to prefer this

Choose this endpoint when you need a quick, low-cost way to bootstrap a small set of synthetic training examples from a known reference text without deploying your own model. It is best suited for early-stage dataset creation, rapid prototyping of fine-tuning pipelines, or generating draft QA pairs for human review. Prefer alternatives when you need verified, high-quality training data at scale, or when output accuracy is critical without manual review.

## Known failure modes

- Task or reference text exceeds character limits (task >300 chars, reference >1000 chars)
- Empty or missing required fields returns validation error
- Low-quality or incoherent output if reference text is ambiguous or poorly structured
- Local model may produce hallucinated or inconsistent completions requiring review
- Payment failure via x402 protocol prevents response delivery

## How this service works

AI Synthetic Training Examples: Generate three short prompt/completion drafts from a task description and supplied reference. Local small-model output, explicitly unverified and requiring review. Local processing. Limits and retention: /ai-services.

## Output

Returns three short prompt/completion draft pairs generated by a local small language model based on the supplied task description and reference text. Output is explicitly unverified and requires human review before use in training pipelines. Results reflect local model inference and may vary in quality.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "task": {
   "type": "string",
   "default": "Draft customer questions and short answers about the refund policy.",
   "maxLength": 300,
   "minLength": 1
  },
  "reference": {
   "type": "string",
   "default": "Northwind support opens Monday to Friday at 09:00 and closes at 17:00. Customers can request a refund within thirty days.",
   "maxLength": 1000,
   "minLength": 1
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/ai-synthetic-training-examples-generator-3637bb77/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from market.datapackvibe.com](https://www.zero.xyz/host/market.datapackvibe.com/llms.txt)