neoapi-sdk 0.1.1

Integrate neoapi.ai LLM Analytics with your LLM pipelines.

NeoAPI SDK

The official Python SDK for integrating neoapi.ai LLM Analytics with your LLM pipelines. Monitor, analyze, and optimize your Language Model outputs with both asynchronous and synchronous implementations.

Installation

pip install neoapi-sdk

Quick Start

from neoapi import NeoApiClient, track_llm_output

# Initialize the client
client = NeoApiClient(api_key="your-api-key")  # Or use NEOAPI_API_KEY env variable

# Track LLM outputs with the decorator
@track_llm_output(
    client=client,
    project="my_project",
    group="my_group"
)
async def my_llm_function():
    return "LLM generated text"

# Start the client before use
await client.start()

# Use your function
result = await my_llm_function()

# Clean up when done
await client.stop()

Key Features

• Asynchronous and synchronous client implementations
• Automatic batching with configurable retry mechanisms
• Dynamic performance optimization
• Simple decorator-based tracking
• Robust error handling
• Flexible metadata support

Configuration

Client Options

client = NeoApiClient(
    api_key="your-api-key",      # Optional if NEOAPI_API_KEY is set
    check_frequency=10,          # Send every Nth LLM output
)

Environment Variables

• NEOAPI_API_KEY: Your NeoAPI API key (allows client initialization without api_key parameter)
• NEOAPI_API_URL: Optional custom API endpoint URL

Advanced Usage

@track_llm_output(
    client=client,
    project="my_project",        # Organize outputs by project
    group="my_group",           # Group related outputs
    analysis_slug="my_analysis", # Unique analysis identifier
    need_analysis_response=True, # Receive analysis results
    format_json_output=True,     # Pretty-print JSON responses
    metadata={"key": "value"}    # Add custom tracking data
)

License

MIT License

Resources

• Documentation: https://www.neoapi.ai/docs
• Source Code: https://github.com/neoapi-ai/neoapi-python
• Support: [e-mail]