GPT-5.4 - Starrise AI

GPT-5.4 is available through Starrise AI via OpenAI Responses API (/v1/responses).

Key Capabilities

Responses API — Uses the new /v1/responses endpoint, replacing messages with input
Reasoning control — Configurable reasoning effort (minimal / low / medium / high) and summary
Verbosity control — Set response verbosity to low, medium, or high
Streaming — Real-time token streaming via SSE
Tool calling — Supports function calling and tool use

Quick Example

curl https://ai.alad.com/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "input": [
      { "role": "user", "content": "Explain quantum entanglement in simple terms." }
    ],
    "reasoning": {
      "effort": "high",
      "summary": "auto"
    },
    "text": {
      "format": { "type": "text" },
      "verbosity": "medium"
    },
    "store": true
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://ai.alad.com/v1"
)

response = client.responses.create(
    model="gpt-5.4",
    input=[
        {"role": "user", "content": "Explain quantum entanglement in simple terms."}
    ],
    reasoning={"effort": "high", "summary": "auto"},
    text={"format": {"type": "text"}, "verbosity": "medium"},
    store=True
)

print(response.output[0].content[0].text)

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://ai.alad.com/v1"
)

stream = client.responses.create(
    model="gpt-5.4",
    input=[
        {"role": "user", "content": "Write a short poem about the ocean."}
    ],
    stream=True
)

for event in stream:
    if hasattr(event, 'delta'):
        print(event.delta, end="")

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Fixed value: `gpt-5.4`
`input`	array	Yes	Array of `{ role, content }` objects
`stream`	boolean	No	Enable SSE streaming, default `false`
`temperature`	float	No	`0`–`2`, controls randomness, default `1`
`top_p`	float	No	Nucleus sampling threshold, default `1`
`max_output_tokens`	integer	No	Maximum output tokens
`stop`	string / array	No	Sequences where generation stops
`reasoning`	object	No	`{ effort, summary }` — Controls reasoning depth
`text`	object	No	`{ format, verbosity }` — Controls output format and verbosity
`tools`	array	No	List of tools the model can call
`store`	boolean	No	Store response for later retrieval, default `true`

API Reference

View the interactive API Playground for GPT-5.4.