Fenra Python SDK

FinOps platform for AI cost visibility

Usage

Install the package:

pip install fenra

Initialize and start tracking:

import fenra
from openai import OpenAI

fenra.init(api_key="your-fenra-api-key")

# Use as normal - costs are tracked automatically
client = OpenAI()
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)

Works with Anthropic (Claude) and Google (Gemini) the same way.

Set context for cost attribution:

fenra.set_context(
    billable_customer_id="cust_123",  # recommended
    environment="production",
    feature="chat",
)

For custom providers or manual tracking:

fenra.track(
    provider="custom",
    model="my-model",
    input_tokens=100,
    output_tokens=50,
)

To disable auto-tracking:

export FENRA_DISABLE_AUTO_TRACK=1

To enable debug logging:

import logging
logging.getLogger("fenra").setLevel(logging.DEBUG)

Support

• Documentation
• Contact

Privacy

Fenra is privacy-first by default. No prompts or responses are stored unless explicitly enabled.

License

The Fenra Python SDK is licensed under the Apache 2.0 License.