Responses (OpenAI)
POST /v1/responses — OpenAI Responses API, parameter by parameter
OpenAI's newer Responses API (/v1/responses); the platform converts it internally to a Chat Completion through the same routing pipeline. Use it when your code targets the OpenAI SDK's responses interface.
Example
from openai import OpenAI
client = OpenAI(base_url="https://api.modelsite.ai/v1", api_key="ms_live_sk_xxx")
r = client.responses.create(
model="claude-sonnet-4.6",
input="Explain quantum entanglement in one sentence",
instructions="Be concise.",
)
print(r.output_text)The API card below is rendered from the OpenAPI spec (Request-body parameter table, Response schema).
Authorization
BearerAuth AuthorizationBearer <token>
Authorization: Bearer $MODELSITE_API_KEY
In: header
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
curl -X POST "https://example.com/responses" \ -H "Content-Type: application/json" \ -d '{ "model": "string", "input": "string" }'{ "id": "string", "object": "string", "model": "string", "output": [ { "type": "string", "id": "string", "status": "string", "role": "string", "content": [ { "type": "string", "text": "string" } ] } ], "status": "completed", "usage": { "input_tokens": 0, "output_tokens": 0, "total_tokens": 0 }}Other methods
| Method | Path | Description |
|---|---|---|
| GET | /v1/responses | List responses |
| GET | /v1/responses/:id | Get response details |
| POST | /v1/responses/:id/cancel | Cancel an in-progress response |
For most cases Chat Completions is enough; use this endpoint only when existing code depends on the Responses API (built-in tools, context continuation).