🚀 Rax AI Python SDK

Official Python SDK for Rax AI Platform - Simple, powerful, and integrated.

Install

pip install rax-ai

Quick Start

fromrax_aiimportRaxAIrax=RaxAI(api_key="your-api-key")
response=rax.chat(
model="rax-4.0",
messages=[
{"role": "user", "content": "Hello!"}
]
)
print(response["choices"][0]["message"]["content"])

Features

• ✅ Simple API - Clean, intuitive methods
• ✅ Python 3.7+ - Compatible with most Python versions
• ✅ Auto Retry - Smart error handling with exponential backoff
• ✅ Type Hints - Full typing support
• ✅ Models API - List and manage available models
• ✅ Usage Tracking - Monitor your API consumption
• ✅ Streaming - Real-time response streaming (coming soon)

Configuration

fromrax_aiimportRaxAIrax=RaxAI(
api_key="your-key",
base_url="https://ai.raxcore.dev/api", # defaulttimeout=30# seconds, default
)

Core Methods

Chat Completions

response=rax.chat(
model="rax-4.0",
messages=[
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": "Explain quantum computing"}
],
max_tokens=150,
temperature=0.7,
top_p=0.9
)
print(response["choices"][0]["message"]["content"])

Streaming Chat

forchunkinrax.chat_stream(
model="rax-4.0",
messages=[{"role": "user", "content": "Tell me a story..."}]
):
print(chunk, end="", flush=True)

Models Management

# Get available modelsmodels=rax.get_models()
print("Available models:", [m["id"] forminmodels["data"]])
# Validate API keyis_valid=rax.validate_key()
print("Key valid:", is_valid)

Usage Analytics

# Get current usageusage=rax.get_usage()
print("Total requests:", usage["total_requests"])
print("Total tokens:", usage["total_tokens"])
# Get usage for date rangemonthly_usage=rax.get_usage(
start_date="2024-01-01", end_date="2024-01-31"
)
print("Monthly breakdown:", monthly_usage["daily_breakdown"])

Error Handling

fromrax_aiimportRaxAI, RaxAIErrortry:
response=rax.chat(
model="rax-4.0",
messages=[{"role": "user", "content": "Hello!"}]
)
exceptRaxAIErrorase:
print(f"API Error: {e.message}")
print(f"Status: {e.status_code}")
print(f"Type: {e.error_type}")
print(f"Details: {e.to_dict()}")

Environment Variables

# .env
RAX_AI_API_KEY=your-key-here

importosfromrax_aiimportRaxAIrax=RaxAI(api_key=os.getenv("RAX_AI_API_KEY"))

Framework Integration

FastAPI

fromfastapiimportFastAPIfromrax_aiimportRaxAIapp=FastAPI()
rax=RaxAI(api_key=os.getenv("RAX_AI_API_KEY"))
@app.post("/chat")asyncdefchat_endpoint(message: str):
response=rax.chat(
model="rax-4.0",
messages=[{"role": "user", "content": message}]
)
returnresponse

Flask

fromflaskimportFlask, request, jsonifyfromrax_aiimportRaxAIapp=Flask(__name__)
rax=RaxAI(api_key=os.getenv("RAX_AI_API_KEY"))
@app.route("/chat", methods=["POST"])defchat():
message=request.json["message"]
response=rax.chat(
model="rax-4.0",
messages=[{"role": "user", "content": message}]
)
returnjsonify(response)

Django

# views.pyfromdjango.httpimportJsonResponsefromrax_aiimportRaxAIimportjsonrax=RaxAI(api_key=settings.RAX_AI_API_KEY)
defchat_view(request):
data=json.loads(request.body)
response=rax.chat(
model="rax-4.0",
messages=[{"role": "user", "content": data["message"]}]
)
returnJsonResponse(response)

Advanced Usage

Configuration Management

# Get current configconfig=rax.get_config()
print("Base URL:", config["base_url"])
print("Timeout:", config["timeout"])

Retry Logic

• 3 automatic retries for network errors
• Exponential backoff (1s, 2s, 4s delays)
• Smart error classification (no retry for 4xx errors)
• Timeout handling with configurable timeouts

Python Version Support

• ✅ Python 3.7+
• ✅ Python 3.8+
• ✅ Python 3.9+
• ✅ Python 3.10+
• ✅ Python 3.11+
• ✅ Python 3.12+

Get Your API Key

1. Visit ai.raxcore.dev
2. Sign up or log in
3. Navigate to API Keys
4. Create a new key
5. Start building!

Type Hints

Full type hints included for better IDE support:

fromrax_ai.typesimportChatMessage, ChatRequest, ChatResponsemessages: List[ChatMessage] = [
{"role": "user", "content": "Hello!"}
]
response: ChatResponse=rax.chat(
model="rax-4.0",
messages=messages
)

Support

• 📖 Documentation
• 🎮 Playground
• 📊 Dashboard
• 🐛 Issues

---

Built for the Rax AI Platform at ai.raxcore.dev