OpenRouter-style API for AI model access. Self-hosted models at a fraction of the cost. Compatible with OpenAI's chat completions API.
curl -X POST https://api.nebulahq.work/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "qwen3:latest", "messages": [{"role": "user", "content": "Hello"}]}'
API keys and authorization
Send messages and get AI responses
Available models and pricing
Token-based pricing
All API requests require an API key passed in the Authorization header:
Authorization: Bearer YOUR_API_KEYGet your API key from the dashboard.
/v1/chat/completions
Send a chat conversation and get a completion response. Compatible with OpenAI's format.
{
"model": "qwen3:latest", // Required: model name
"messages": [ // Required: conversation
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Hello!"}
],
"temperature": 0.7, // Optional: 0.0-2.0, default 0.7
"max_tokens": 2048, // Optional: max response tokens
"stream": false // Optional: for streaming responses
}{
"id": "chatcmpl-abc123",
"object": "chat.completion",
"created": 1712531200,
"model": "qwen3:latest",
"choices": [{
"index": 0,
"message": {
"role": "assistant",
"content": "Hello! How can I help you today?"
},
"finish_reason": "stop"
}],
"usage": {
"prompt_tokens": 10,
"completion_tokens": 12,
"total_tokens": 22
}
}| Model | Context | Use Case | Input/1M | Output/1M |
|---|---|---|---|---|
| qwen3:latest | 32K | General, reasoning | $0.20 | $0.60 |
| gemma3:1b | 4K | Fast, lightweight | $0.05 | $0.10 |
| qwen3-vl:235b-cloud | 32K | Vision, multimodal | $0.50 | $1.50 |
| deepseek-v3.1:671b-cloud | 64K | Advanced reasoning | $0.35 | $0.90 |
| qwen/qwen3-coder:free | 32K | Code generation (free) | Free | Free |
import requests
API_KEY = "YOUR_API_KEY"
BASE_URL = "https://api.nebulahq.work"
headers = {
"Authorization": f"Bearer {API_KEY}",
"Content-Type": "application/json"
}
response = requests.post(
f"{BASE_URL}/v1/chat/completions",
headers=headers,
json={
"model": "qwen3:latest",
"messages": [{"role": "user", "content": "Hello!"}]
}
)
print(response.json()["choices"][0]["message"]["content"])const response = await fetch('https://api.nebulahq.work/v1/chat/completions', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({
model: 'qwen3:latest',
messages: [{ role: 'user', content: 'Hello!' }]
})
});
const data = await response.json();
console.log(data.choices[0].message.content);