GPT-5.4 nano · Chat completions

POST/v1/chat/completions

Request parameters

Authorization

Add an Authorization header whose value is Bearer followed by your API key.

Authorization: Bearer ****************

Header parameters

Content-TypestringOptional

application/json

AcceptstringOptional

application/json

Body parametersapplication/jsonrequired

modelstringRequired

Model ID, fixed to `gpt-5.4-nano`

messagesarray[object]Required

List of conversation messages; each has `role` (system / user / assistant) and `content`

streamboolOptional

Whether to stream the response (SSE), default `false`

temperaturenumberOptional

Sampling temperature; higher is more random

top_pnumberOptional

Nucleus sampling; alternative to `temperature`. Range `0`–`1`

max_tokensintOptional

Maximum number of generated tokens

nintOptional

Number of completions to generate, default `1`

stoparrayOptional

Stop sequences; output halts when any is produced

presence_penaltynumberOptional

Penalize tokens already present to boost novelty. Range `-2`–`2`

frequency_penaltynumberOptional

Penalize tokens by frequency to reduce repetition. Range `-2`–`2`

logprobsboolOptional

Whether to return log probabilities of the output tokens

top_logprobsintOptional

Number of most likely tokens to return per position (requires `logprobs`)

seedintOptional

Random seed for best-effort reproducible sampling

reasoning_effortstringOptional

Reasoning depth for thinking models: `low` / `medium` / `high`

toolsarrayOptional

Tool / function definitions for function calling

tool_choicestring/objectOptional

Whether and which tool to call: `auto` / `none` / a named tool

parallel_tool_callsboolOptional

Whether to allow calling multiple tools in parallel

response_formatobjectOptional

Constrain the output format, e.g. force a JSON object

userstringOptional

End-user identifier, used for abuse monitoring

Example

{
  "model": "gpt-5.4-nano",
  "messages": [
    {
      "role": "user",
      "content": "Hi, tell me about yourself"
    }
  ]
}

{  "model": "gpt-5.4-nano",  "messages": [    {      "role": "user",      "content": "Hi, tell me about yourself"    }  ]}

Request code samples

curl https://api.hotrouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4-nano",
    "messages": [
      {
        "role": "user",
        "content": "Hi, tell me about yourself"
      }
    ]
  }'

curl https://api.hotrouter.ai/v1/chat/completions \  -H "Authorization: Bearer $API_KEY" \  -H "Content-Type: application/json" \  -d '{    "model": "gpt-5.4-nano",    "messages": [      {        "role": "user",        "content": "Hi, tell me about yourself"      }    ]  }'

Response

200 · Success

idstringRequired

Unique id of this response

objectstringRequired

Object type

createdintegerRequired

Creation timestamp (seconds)

modelstringRequired

Model actually used

choicesarray[object]Required

List of generated results

usageobjectRequired

Token usage of this call

Example

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1730000000,
  "model": "gpt-5.4-nano",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hi! Nice to meet you…"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 24,
    "total_tokens": 36
  }
}

{  "id": "chatcmpl-xxx",  "object": "chat.completion",  "created": 1730000000,  "model": "gpt-5.4-nano",  "choices": [    {      "index": 0,      "message": {        "role": "assistant",        "content": "Hi! Nice to meet you…"      },      "finish_reason": "stop"    }  ],  "usage": {    "prompt_tokens": 12,    "completion_tokens": 24,    "total_tokens": 36  }}

GPT-5.4 nano · Chat completions

Request parameters

Request code samples

Response

On this page