---
title: Responses API
description: Generate a response through the OpenAI-compatible Responses API.
---

<ApiEndpoint method="POST" path="/v1/responses" auth="api_key" />

Use the Responses API when your OpenAI client or application already uses `responses.create`. Zorveus forwards supported request fields to the selected provider route and records the resulting usage.

<KeyInserter type="api" label="API Key Tester" placeholder="zrv_your_api_key" />

---

## Code example

<CodeGroup titles={["Python (OpenAI SDK)", "cURL"]}>

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.zorveus.com/v1",
    api_key="zrv_your_api_key",
)

response = client.responses.create(
    model="openai/gpt-4.1-mini",
    input="What are three core benefits of edge computing?",
)

print(response.output_text)
```

```bash
curl https://api.zorveus.com/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer zrv_your_api_key" \
  -d '{
    "model": "openai/gpt-4.1-mini",
    "input": "What are three core benefits of edge computing?"
  }'
```

</CodeGroup>

## Expected result

A successful response contains output items and, when the SDK can derive it, `output_text`. The exact item types depend on the model and requested tools.

Use `GET /v1/models` before selecting a model. Route availability does not prove that every model supports every Responses API feature.

## Common failures

- `zorveus_model_not_allowed` means the API key policy excludes the model.
- `zorveus_unsupported_model` means no current route can serve it.
- Wallet, cap, or product-credit errors occur before provider spend.