# ManyLives Quick AI Inference

> ManyLives Quick AI Inference is a paid API for AI agents from api.manylives.me, paid per call via x402, $0.001/call, status unknown (last checked 2026-10-12).

Runs a fast, cheap LLM inference call (Llama 3.1 8B) for a given prompt, returning a text reply with usage stats, billed per call at $0.001 USDC with no account required.

## Facts

- Endpoint: POST https://api.manylives.me/paid/ai/quick?utm_source=zero.xyz
- Price: $0.001/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/manylives-quick-ai-inference-b4cab7c3
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_kLRS9I1FAlYUMr1Jefv8V

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 manylives-quick-ai-inference-b4cab7c3 -d '<json body>'
```

Example prompt: Classify the sentiment of this customer review as positive, negative, or neutral using the ManyLives quick AI endpoint: 'The delivery was fast and the product was exactly what I ordered.'

## When to prefer this

Choose this endpoint when you need a lightweight, pay-per-call LLM inference with no subscription, no account setup, and ultra-low cost ($0.001 per call). Ideal for agents that run occasional or unpredictable workloads where a monthly API subscription is wasteful. Best for short classification, extraction, or summarization tasks fitting within the 200-token default output window. Not suited for long-form generation or tasks requiring GPT-4-class reasoning.

## Known failure modes

- Prompt too long or malformed — 400 error
- max_tokens out of range (must be 1–256) — validation error
- temperature out of range (must be 0–1.5) — validation error
- Payment not provided or insufficient — 402 Payment Required
- Model timeout on complex prompts — 5xx error
- Response truncated if reply exceeds max_tokens limit

## How this service works

Pay-per-call tools for people building AI agents: live web search and page contents, point-in-time crypto market data and news for backtests (no lookahead), verifiable track records of 1,000 crypto strategies locked on the Base network, company profiles, quick AI and batch prices. Ask us to turn a public website into a JSON endpoint. Pay per call, no account, free trial.

## Output

Returns a JSON object containing the model name (e.g. 'llama-3.1-8b'), the text reply from the model, and a usage object showing the max_tokens setting and input character count.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "prompt": {
   "type": "string",
   "examples": [
    "Classify the sentiment of: The delivery was fast and exactly right."
   ],
   "description": "the request (or messages: [{role, content}])"
  },
  "max_tokens": {
   "type": "string",
   "description": "optional 1-256 (default 200)"
  },
  "temperature": {
   "type": "string",
   "description": "optional 0-1.5"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "model": "llama-3.1-8b",
  "reply": "positive",
  "usage": {
   "max_tokens": 200,
   "input_chars": 64
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/manylives-quick-ai-inference-b4cab7c3/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.manylives.me](https://www.zero.xyz/host/api.manylives.me/llms.txt)