Quick Start

Get started with Nebul's Private inference-api in minutes. The Private inference-api provides instant access to the latest language and other AI models, including Mistral Large 3, OpenAI's GPT-OSS, DeepSeek R1, Qwen3, and more, all through a single OpenAI-compatible interface. Deploy AI capabilities without the complexity of managing infrastructure, while maintaining full privacy and security for your data.

Prerequisites

Before you begin, you'll need:

• A Nebul AI Studio with API access
• An API key (get one from your Nebul AI Studio)
• If using Python, you need to use Python version 3.9 or higher

Installation

Python

bash

1

pip install openai

cURL

No installation required - cURL is typically pre-installed on most systems.

The Nebul inference-api is fully compatible with the OpenAI SDK, so you can use the same library you're already familiar with.

Available Endpoints

Endpoint	Method	Description
/v1/chat/completions	POST	Chat-based completions (recommended)
/v1/completions	POST	Legacy text completions
/v1/responses	POST	Advanced responses with async support
/v1/embeddings	POST	Generate vector embeddings
/v1/rerank	POST	Rerank documents by relevance
/v1/models	GET	List available models
/v1/audio/transcriptions	POST	Speech to text
/v1/audio/speech	POST	Text to speech
/v1/ocr	POST	Optical character recognition
/v1/images/generations	POST	Image generation
/v1/images/edits	POST	Image editing
/v1/images/variations	POST	Image variations

List Available Models

Discover which models are available to your account:

Python

python

12345678910

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key-here",
    base_url="https://api.inference.nebul.io/v1"
)

models = client.models.list()
for model in models.data:
    print(model.id)

cURL

bash

12

curl https://api.inference.nebul.io/v1/models \
  -H "Authorization: Bearer sk-your-api-key-here"

Your First LLM Request

Here's a simple example to get you started:

Python

python

12345678910111213141516

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key-here",
    base_url="https://api.inference.nebul.io/v1"
)

# Replace openai/gpt-oss-120b with the model you want to use returned by the models endpoint
response = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[
        {"role": "user", "content": "Why is privacy important in the age of AI?"}
    ]
)

print(response.choices[0].message.content)

cURL

bash

123456789101112

curl https://api.inference.nebul.io/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-api-key-here" \
  -d '{
    "model": "openai/gpt-oss-120b",
    "messages": [
      {
        "role": "user",
        "content": "Why is privacy important in the age of AI?"
      }
    ]
  }'

Replace sk-your-api-key-here with your actual API key from Nebul AI Studio.

tip

API keys always start with sk- ("secret key-"). Store them securely in environment variables rather than hardcoding.

Error Handling

API errors follow a standard format:

json

12345678

{
  "error": {
    "message": "Invalid API key provided",
    "type": "auth_error",
    "param": null,
    "code": "401"
  }
}

Common error codes:

Code	Meaning
401	Invalid or missing API key
403	Model not available for your account
429	Rate limit exceeded (see limits)
500	Server error — retry the request

What's Next?

Now that you've made your first request, explore more capabilities:

• Examples - See more code examples and use cases
• Models - Browse available models and their capabilities
• API Reference - Explore the complete API documentation