# GPT Trial Endpoint

> GPT Trial Endpoint is a paid API for AI agents from gpt.558686.xyz, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-02).

Provides a pay-per-call GPT-based AI text generation or task completion trial via a simple prompt-in, response-out interface.

## Facts

- Endpoint: GET https://gpt.558686.xyz/v1/trial?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gpt-trial-endpoint-49f594e3
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_hEDLlw4MxFIiAR6BQW2IT

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 gpt-trial-endpoint-49f594e3
```

Example prompt: Can you run this through the GPT trial endpoint and get me a short summary of this paragraph: 'Artificial intelligence is transforming industries by automating repetitive tasks, enabling smarter decision-making, and opening new possibilities for human creativity.'

## When to prefer this

Choose this endpoint when you need a low-cost, one-off GPT text generation call without committing to a subscription, and your use case fits a simple prompt-in/text-out pattern. It is well-suited for trial integrations, quick automation steps, or lightweight agent tasks where $0.01 per call is acceptable.

## Known failure modes

- Missing or empty 'input' field returns a 400 or error response
- Insufficient USDC payment causes 402 Payment Required error
- Ambiguous or malformed prompt may yield low-quality or irrelevant completions
- Rate limiting or backend GPT API errors may cause 500 responses
- No schema validation on 'params' may silently ignore unsupported options

## Output

Returns an AI-generated text response to the submitted prompt or task, likely as a plain text or JSON payload containing the model's completion output.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "prompt": {
   "type": "string",
   "maxLength": 500,
   "description": "Your prompt (max 500 characters)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gpt-trial-endpoint-49f594e3/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from gpt.558686.xyz](https://www.zero.xyz/host/gpt.558686.xyz/llms.txt)