# Losbeto /classify — Zero-Shot Text Classification

> Losbeto /classify — Zero-Shot Text Classification is a paid API for AI agents from markets.losbeto.xyz, paid per call via x402, $0.008/call, status unknown (last checked 2026-10-14).

Classifies input text into exactly one label from a caller-supplied closed set, returning the matched label verbatim and a 0–1 confidence score, paid per-call in USDC via x402.

## Facts

- Endpoint: POST https://markets.losbeto.xyz/classify?utm_source=zero.xyz
- Price: $0.008/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/losbeto-classify-zero-shot-text-classification-6476836f
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_7-SqsvpoX12-Qzdk1BRH5

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 losbeto-classify-zero-shot-text-classification-6476836f -d '<json body>'
```

Example prompt: Classify this support message into exactly one of these labels — 'billing', 'technical issue', 'feature request', 'general inquiry' — and give me a confidence score: 'I was charged twice for my subscription this month and I need a refund immediately.'

## When to prefer this

Use this endpoint when you need deterministic, closed-set text classification where the answer must be exactly one of your labels — support triage, intent routing, spam filtering, or sentiment bucketing. Unlike open-ended extraction or tagging, this endpoint guarantees the output is one of your supplied labels verbatim. Prefer it over general LLM chat when you need structured, programmatic classification output without managing API keys or monthly subscriptions, paying only per successful call in USDC.

## Known failure modes

- Model returns a label not in the provided list — payment refunded as call credit
- Fewer than 2 or more than 30 labels provided — validation error
- Text exceeds 6000 character limit — rejected
- Payment in unsupported asset or chain — x402 payment failure
- Network or gateway timeout — call fails with no charge

## How this service works

OpenAI-compatible LLM gateway at $0.005/call — change base_url, pay in USDC. Plus pay-per-call market data: multi-oracle price consensus, sentiment, forex, equities, commodities, Brazil macro (BCB/B3) and crypto. No API keys, no accounts. USDC on Base, Solana, Algorand and Polygon via x402.

## Output

A JSON object containing the chosen label (copied verbatim from the caller's candidate list) and a confidence score between 0 and 1. If the model's answer falls outside the provided label set, the payment is returned as call credit.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "properties": {
    "method": {
     "type": "string",
     "description": "POST"
    },
    "bodyParams": {
     "type": "object",
     "required": [
      "text",
      "labels"
     ],
     "properties": {
      "text": {
       "type": "string",
       "description": "Text to classify (max 6000 chars)."
      },
      "labels": {
       "type": "array",
       "items": {
        "type": "string"
       },
       "description": "2-30 candidate labels; the answer is always one of them."
      }
     }
    }
   }
  },
  "output": {
   "type": "object",
   "properties": {
    "example": {
     "type": "object",
     "description": "Assign text to exactly one label from a closed set you provide — support triage, intent routing, spam/not-spam, sentiment buckets. The answer is always one of your labels, copied verbatim, with a 0-1 confidence; if the model returns anything else the payment is returned as call credit. Do NOT use for open-ended tagging — that is /extract."
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/losbeto-classify-zero-shot-text-classification-6476836f/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from markets.losbeto.xyz](https://www.zero.xyz/host/markets.losbeto.xyz/llms.txt)