The Serverspace control panel provides a wide range of modern artificial intelligence models from different providers. Before sending your first request, it is important to choose the appropriate model depending on the task — speed, response quality, context window, and cost.
Available models at the moment
Parameter configuration and request execution
After selecting a model, a menu will open where you can manage API keys and model parameters to achieve the best results.
The following settings are available:
- API key selection — by default, the default key is used
- Model selection — for example, Claude Haiku 4.5 or GPT-5.4
- Temperature — level of response creativity (e.g. 0.6)
- Top P — probability sampling parameter (e.g. 0.1)
- Maximum tokens — limits the response length
API endpoint
The following endpoint is used for sending requests:
Example of the first request
To make a request via the GPT API, you need to form an HTTP request using the selected API key and model. The cURL format is commonly used:
-H "Content-Type: application/json"
-H "Authorization: Bearer YOUR_API_KEY"
-d '{
"model": "gpt-5.4-mini",
"messages": [
{
"role": "user",
"content": "Hello! Explain how the GPT API works."
}
],
"temperature": 0.6,
"top_p": 0.1,
"max_tokens": 500
}'
After executing the request, you will receive the model response in JSON format, which can be used in your application or service.