SDKs & Libraries
Use our official SDKs to integrate Fusion AI into your applications with just a few lines of code.
Python SDK
pip install fusion-ai
JavaScript SDK
npm install @fusion-ai/sdk
Postman Collection
Import & test instantly
cURL Examples
Command line testing
Python SDK
Installation
pip install fusion-ai
Basic Usage
from fusion_ai import FusionAI
# Initialize the client
client = FusionAI(api_key="sk-fusion-your-key-here")
# Simple chat completion
response = client.chat(
prompt="Explain machine learning in simple terms",
provider="neuroswitch" # Let NeuroSwitch choose the best model
)
print(response.text)
print(f"Used: {response.provider}/{response.model}")
print(f"Tokens: {response.tokens.total_tokens}")Advanced Examples
Streaming Response
for chunk in client.chat_stream(
prompt="Write a story about AI",
provider="neuroswitch"
):
print(chunk.text, end="", flush=True)Specific Provider
response = client.chat(
prompt="Code review this function",
provider="claude",
model="claude-3-sonnet"
)With BYOAPI Key
response = client.chat(
prompt="Hello world",
provider="openai",
byoapi_key="your-openai-key"
)Error Handling
try:
response = client.chat(prompt="Hello")
except FusionAIError as e:
print(f"Error: {e.message}")
print(f"Code: {e.status_code}")JavaScript SDK
Installation
npm install @fusion-ai/sdk
Basic Usage
import FusionAI from '@fusion-ai/sdk';
// Initialize the client
const client = new FusionAI({
apiKey: 'sk-fusion-your-key-here'
});
// Simple chat completion
const response = await client.chat({
prompt: 'Explain machine learning in simple terms',
provider: 'neuroswitch'
});
console.log(response.text);
console.log(`Used: ${response.provider}/${response.model}`);
console.log(`Tokens: ${response.tokens.total_tokens}`);Advanced Examples
Streaming Response
for await (const chunk of client.chatStream({
prompt: 'Write a story about AI',
provider: 'neuroswitch'
})) {
process.stdout.write(chunk.text);
}React Hook
import { useFusionAI } from '@fusion-ai/react';
const { chat, loading, error } = useFusionAI();
const handleSubmit = async () => {
const response = await chat({
prompt: userInput,
provider: 'neuroswitch'
});
};Express.js Route
app.post('/api/chat', async (req, res) => {
try {
const response = await client.chat({
prompt: req.body.prompt,
provider: 'neuroswitch'
});
res.json(response);
} catch (error) {
res.status(500).json({ error: error.message });
}
});TypeScript Types
interface ChatRequest {
prompt: string;
provider?: Provider;
model?: string;
temperature?: number;
max_tokens?: number;
}Postman Collection
Import our pre-configured Postman collection to start testing the Fusion AI API immediately.
What's Included
- Chat completion examples
- All provider options (NeuroSwitch, OpenAI, Claude, Gemini)
- Streaming requests
- BYOAPI examples
- Error handling scenarios
- Environment variables setup
cURL Examples
Basic Chat Request
curl -X POST https://api.mcp4.ai/chat \
-H "Authorization: Bearer sk-fusion-your-key-here" \
-H "Content-Type: application/json" \
-d '{
"prompt": "Hello, how are you?",
"provider": "neuroswitch"
}'Streaming Request
curl -X POST https://api.mcp4.ai/chat \
-H "Authorization: Bearer sk-fusion-..." \
-H "Content-Type: application/json" \
-d '{
"prompt": "Write a story",
"provider": "neuroswitch",
"stream": true
}'Specific Provider
curl -X POST https://api.mcp4.ai/chat \
-H "Authorization: Bearer sk-fusion-..." \
-H "Content-Type: application/json" \
-d '{
"prompt": "Explain quantum physics",
"provider": "claude",
"model": "claude-3-sonnet"
}'With Parameters
curl -X POST https://api.mcp4.ai/chat \
-H "Authorization: Bearer sk-fusion-..." \
-H "Content-Type: application/json" \
-d '{
"prompt": "Be creative",
"provider": "neuroswitch",
"temperature": 0.9,
"max_tokens": 500
}'Check Credits
curl -H "Authorization: Bearer sk-fusion-..." \ https://api.mcp4.ai/credits