# BlockRun Hosted GPT-4o / GPT-4.1 Text Chat

> BlockRun Hosted GPT-4o / GPT-4.1 Text Chat is a paid API for AI agents from market.datapackvibe.com, paid per call via x402, $0.09/call, status unknown (last checked 2026-10-27).

Provides pay-per-call hosted text chat completions using OpenAI GPT-4o or GPT-4.1, with up to 8 conversation turns and 2,048 output tokens, returning the answer, served model name, and token usage.

## Facts

- Endpoint: POST https://market.datapackvibe.com/x402/assistant-hosted-premium-chat?utm_source=zero.xyz
- Price: $0.09/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-27
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/blockrun-hosted-gpt-4o-gpt-4-1-text-chat-054271e7
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_SFqP9dmnEnM9elyLO8BWB

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 blockrun-hosted-gpt-4o-gpt-4-1-text-chat-054271e7 -d '<json body>'
```

Example prompt: Ask GPT-4o via the BlockRun hosted chat endpoint: 'What are the key differences between REST and GraphQL?' — use up to 512 output tokens and keep the system message as 'You are a concise technical assistant.'

## When to prefer this

Choose this endpoint when you need a quick, bounded, pay-per-call GPT-4o or GPT-4.1 text completion without managing your own OpenAI API key or subscription — especially in agent pipelines where cost control per call matters (fixed $0.09 USDC per call). Prefer this over self-hosted or subscription LLMs when you want metered usage with no monthly commitment. Not suitable when you need vision inputs, function/tool calling, streaming responses, or more than 2,048 output tokens.

## Known failure modes

- Input exceeds 16,000 character limit — request rejected with validation error
- Message array exceeds 8 turns — request rejected
- max_tokens outside 1–2048 range — validation error
- Invalid model enum value — rejected with schema error
- Payment not received or x402 protocol failure — 402 response, no completion returned
- Provider (OpenAI) downstream error — may return 5xx or error message
- Empty or missing messages array — validation error

## How this service works

Hosted GPT-4o / GPT-4.1 Text Chat: Hosted GPT-4o or GPT-4.1 text chat through BlockRun. Up to 16,000 total input characters, eight turns and 2,048 output tokens. Returns answer, served model and usage. Inputs leave this PC for the named providers. Text only; no vision, tools or streaming. 0.09 USDC includes one bounded hosted text answer.

## Output

Returns a JSON object containing the generated text answer from GPT-4o or GPT-4.1, the name of the model that actually served the request, and token usage statistics (prompt tokens, completion tokens, total tokens).

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "model": {
   "enum": [
    "openai/gpt-4o",
    "openai/gpt-4.1"
   ],
   "type": "string",
   "default": "openai/gpt-4o"
  },
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "user",
       "assistant"
      ],
      "type": "string"
     },
     "content": {
      "type": "string",
      "maxLength": 16000,
      "minLength": 1
     }
    },
    "additionalProperties": false
   },
   "default": [
    {
     "role": "user",
     "content": "Explain HTTP 402 in one sentence."
    }
   ],
   "maxItems": 8,
   "minItems": 1
  },
  "max_tokens": {
   "type": "integer",
   "default": 128,
   "maximum": 2048,
   "minimum": 1
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/blockrun-hosted-gpt-4o-gpt-4-1-text-chat-054271e7/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)