# OpenAI Responses API

> OpenAI Responses API is a paid API for AI agents from OpenAI at openai.mpp.tempo.xyz, paid per call via MPP, $0.005125/call, status unknown (last checked 2026-10-11, last successful call 2026-08-13).

Stateful multi-turn AI responses using OpenAI models (Codex, GPT-4o, etc.) with built-in tool use and conversation management

## Facts

- Endpoint: POST https://openai.mpp.tempo.xyz/v1/responses?utm_source=zero.xyz
- Price: $0.005125/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-11
- Last successful call: 2026-08-13
- Success rate: 40% of calls made through Zero
- Activations on Zero: 6
- Provider: OpenAI
- Docs: https://platform.openai.com/docs
- Website: https://openai.com
- Tags: llm, gpt-4o, dall-e, whisper, tts, embeddings, chat, ai, media
- Canonical page: https://www.zero.xyz/c/openai-responses-57dd30b3
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_gIN_rZyXhMdGo5FJnmDJR

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 openai-responses-57dd30b3 -d '<json body>'
```

Example prompt: Using OpenAI's Responses API with GPT-4o, answer the following question with web search tool enabled: 'What are the top 3 breakthroughs in AI from the past 6 months?' and return a concise summary with sources.

## When to prefer this

Choose this endpoint over the standard Completions API when you need stateful multi-turn conversations, built-in tool use orchestration, or agentic task execution with persistent context. Prefer it for complex reasoning tasks where conversation memory matters. Use Completions for simpler one-shot text generation without state management.

## Known failure modes

- Invalid or unsupported model ID returns 400 Bad Request
- Malformed request body or missing required fields returns 422 Unprocessable Entity
- Insufficient USDC balance or payment failure returns 402 Payment Required
- Rate limit exceeded returns 429 Too Many Requests
- Model overloaded or service unavailable returns 503
- Context window exceeded with very long conversation history returns 400

## How this service works

Responses API (Codex, GPT-4o, etc.) - price varies by model

## Output

Returns a structured response object containing the AI-generated text output, any tool calls made, conversation state for follow-up turns, token usage counts, model used, and finish reason. The response supports multi-turn continuations via stored conversation context.

## Example request

```json
{
 "input": "What is the capital of France?",
 "model": "gpt-4o"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/openai-responses-57dd30b3/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from openai.mpp.tempo.xyz](https://www.zero.xyz/host/openai.mpp.tempo.xyz/llms.txt)