Create a model response

curl --request POST \
  --url https://api.synapsai.cloud/v1/responses \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "input": "<string>",
  "instructions": "<string>",
  "max_output_tokens": 123,
  "max_tool_calls": 123,
  "metadata": {},
  "logit_bias": {},
  "parallel_tool_calls": true,
  "prompt": {},
  "reasoning": {},
  "service_tier": "auto",
  "store": false,
  "stream": false,
  "temperature": 123,
  "text": {},
  "tool_choice": "auto",
  "tools": [
    {}
  ],
  "top_logprobs": 0,
  "top_p": 123,
  "top_k": 123,
  "truncation": "disabled",
  "skip_special_tokens": true,
  "presence_penalty": 123,
  "frequency_penalty": 123,
  "cache_salt": "<string>",
  "repetition_penalty": 123,
  "seed": 123,
  "stop": "<string>",
  "structured_outputs": {},
  "chat_template_kwargs": {},
  "include": [
    "<string>"
  ],
  "previous_input_messages": [
    {}
  ],
  "previous_response_id": "<string>",
  "background": false
}
'

import requests

url = "https://api.synapsai.cloud/v1/responses"

payload = {
    "model": "<string>",
    "input": "<string>",
    "instructions": "<string>",
    "max_output_tokens": 123,
    "max_tool_calls": 123,
    "metadata": {},
    "logit_bias": {},
    "parallel_tool_calls": True,
    "prompt": {},
    "reasoning": {},
    "service_tier": "auto",
    "store": False,
    "stream": False,
    "temperature": 123,
    "text": {},
    "tool_choice": "auto",
    "tools": [{}],
    "top_logprobs": 0,
    "top_p": 123,
    "top_k": 123,
    "truncation": "disabled",
    "skip_special_tokens": True,
    "presence_penalty": 123,
    "frequency_penalty": 123,
    "cache_salt": "<string>",
    "repetition_penalty": 123,
    "seed": 123,
    "stop": "<string>",
    "structured_outputs": {},
    "chat_template_kwargs": {},
    "include": ["<string>"],
    "previous_input_messages": [{}],
    "previous_response_id": "<string>",
    "background": False
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "id": "<string>",
  "object": "response",
  "created_at": 123,
  "model": "<string>",
  "output": [
    {
      "id": "<string>",
      "type": "<string>",
      "role": "<string>",
      "content": [
        {}
      ],
      "call_id": "<string>",
      "name": "<string>",
      "arguments": "<string>"
    }
  ],
  "output_text": "<string>",
  "instructions": "<string>",
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  },
  "error": {
    "code": "<string>",
    "message": "<string>"
  },
  "incomplete_details": {}
}

API Reference

Create a model response

OpenAI-compatible Responses API backed by vLLM deployments only. Provide text or structured input to generate a response with optional tool calling, reasoning, and streaming.

Availability: Only supported on vLLM-backed model deployments.

Stateless operation: Responses are not stored server-side. The store parameter is always forced to false. Use previous_input_messages to carry conversation history across turns instead of previous_response_id.

Unsupported parameters: Requests that set previous_response_id or background: true are rejected at validation with a clear error.

Batching: This endpoint does not support batch requests.

POST

responses

Create a model response

curl --request POST \
  --url https://api.synapsai.cloud/v1/responses \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "input": "<string>",
  "instructions": "<string>",
  "max_output_tokens": 123,
  "max_tool_calls": 123,
  "metadata": {},
  "logit_bias": {},
  "parallel_tool_calls": true,
  "prompt": {},
  "reasoning": {},
  "service_tier": "auto",
  "store": false,
  "stream": false,
  "temperature": 123,
  "text": {},
  "tool_choice": "auto",
  "tools": [
    {}
  ],
  "top_logprobs": 0,
  "top_p": 123,
  "top_k": 123,
  "truncation": "disabled",
  "skip_special_tokens": true,
  "presence_penalty": 123,
  "frequency_penalty": 123,
  "cache_salt": "<string>",
  "repetition_penalty": 123,
  "seed": 123,
  "stop": "<string>",
  "structured_outputs": {},
  "chat_template_kwargs": {},
  "include": [
    "<string>"
  ],
  "previous_input_messages": [
    {}
  ],
  "previous_response_id": "<string>",
  "background": false
}
'

import requests

url = "https://api.synapsai.cloud/v1/responses"

payload = {
    "model": "<string>",
    "input": "<string>",
    "instructions": "<string>",
    "max_output_tokens": 123,
    "max_tool_calls": 123,
    "metadata": {},
    "logit_bias": {},
    "parallel_tool_calls": True,
    "prompt": {},
    "reasoning": {},
    "service_tier": "auto",
    "store": False,
    "stream": False,
    "temperature": 123,
    "text": {},
    "tool_choice": "auto",
    "tools": [{}],
    "top_logprobs": 0,
    "top_p": 123,
    "top_k": 123,
    "truncation": "disabled",
    "skip_special_tokens": True,
    "presence_penalty": 123,
    "frequency_penalty": 123,
    "cache_salt": "<string>",
    "repetition_penalty": 123,
    "seed": 123,
    "stop": "<string>",
    "structured_outputs": {},
    "chat_template_kwargs": {},
    "include": ["<string>"],
    "previous_input_messages": [{}],
    "previous_response_id": "<string>",
    "background": False
}
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "id": "<string>",
  "object": "response",
  "created_at": 123,
  "model": "<string>",
  "output": [
    {
      "id": "<string>",
      "type": "<string>",
      "role": "<string>",
      "content": [
        {}
      ],
      "call_id": "<string>",
      "name": "<string>",
      "arguments": "<string>"
    }
  ],
  "output_text": "<string>",
  "instructions": "<string>",
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  },
  "error": {
    "code": "<string>",
    "message": "<string>"
  },
  "incomplete_details": {}
}

Body

application/json

Request body for the OpenAI-compatible Responses API. The primary input field is input. Only available on vLLM-backed deployments. Responses are not persisted; store is always forced to false.

model

string

required

Model ID to use for this request. Required.

input

required

Text or structured input to the model. Required.

instructions

string | null

System or developer instructions inserted into the model context, separate from input.

max_output_tokens

integer | null

Maximum number of tokens the model may generate in the response.

max_tool_calls

integer | null

Maximum number of tool calls the model may make during generation.

metadata

object | null

Optional key-value metadata attached to the request.

logit_bias

object | null

Mapping from token ID to a bias value added to the token logits.

Show child attributes

parallel_tool_calls

boolean

default:true

Whether the model may call multiple tools in parallel.

prompt

object | null

Optional prompt template configuration.

reasoning

object | null

Optional reasoning configuration for supported models.

service_tier

enum<string>

default:auto

Service tier hint for request scheduling.

Available options:

auto,

default,

flex,

scale,

priority

store

boolean

default:false

Ignored — always forced to false because responses are not stored server-side.

stream

boolean

default:false

If true, stream partial response events as Server-Sent Events (SSE).

temperature

number | null

Sampling temperature; higher values increase randomness.

text

object | null

Text output configuration, including structured output format via text.format.

tool_choice

default:auto

Controls which tool the model calls. Accepts a string (for example, auto, none, required) or an object specifying a function.

tools

object[] | null

Optional list of tool definitions the model may call.

top_logprobs

integer | null

default:0

Return the top N token logprobs at each position when logprobs are enabled.

top_p

number | null

Nucleus sampling cumulative probability threshold.

top_k

integer | null

Keep only the top_k tokens by probability at each generation step.

truncation

enum<string>

default:disabled

Truncation strategy for inputs that exceed the model context window.

Available options:

auto,

disabled

skip_special_tokens

boolean

default:true

Whether to skip special tokens in the generated output.

presence_penalty

number | null

Penalty applied to discourage new token presence.

frequency_penalty

number | null

Penalty applied based on token frequency to reduce repetition.

cache_salt

string | null

Optional salt for prefix caching.

repetition_penalty

number | null

Additional repetition control for generation.

seed

integer | null

Optional random seed for reproducibility.

stop

Stop sequence or list of sequences where generation should halt.

structured_outputs

object | null

Structured output schema configuration.

chat_template_kwargs

object | null

Additional keyword arguments passed to the model chat template.

include

string[] | null

Additional output fields to include in the response.

previous_input_messages

object[] | null

Prior conversation messages to include when continuing a multi-turn exchange without server-side response storage.

previous_response_id

string | null

Not supported. Requests that set this field are rejected with a validation error because responses are not stored.

background

boolean

default:false

Not supported. Requests with background: true are rejected with a validation error because responses are not stored.

Response

Model response

OpenAI-compatible response object returned by the Responses API.

string

required

Unique identifier for this response.

object

string

default:response

required

Object type. Always response.

created_at

integer

required

Unix timestamp (seconds) when the response was created.

model

string

required

Model identifier used to generate the response.

status

enum<string>

required

Status of the response generation.

Available options:

completed,

failed,

in_progress,

incomplete,

cancelled

output

object[]

required

Typed output items produced by the model (for example, messages, function calls, or reasoning items).

Show child attributes

output_text

string | null

Convenience field containing concatenated text output from the response.

instructions

string | null

System or developer instructions used for this response.

usage

object | null

Token usage details for the request.

Show child attributes

error

object | null

Error details when generation fails.

Show child attributes

incomplete_details

object | null

Details about why the response ended before completion.

Show child attributes

Create a text completion

Generate embeddings

Overview

Endpoints

Create a model response

Body

Response