nanomon 0.1.0

Track LLM costs and usage across DSPy program runs

Nanomon

Track LLM costs and usage across DSPy program runs.

Installation

pip install nanomon

Quick Start

import dspy
from nanomon import RunLogger, track_tool
from nanomon.sinks import ApiSink
from nanomon.pricing import JsonPricing

# Setup pricing and sink
pricing = JsonPricing("pricing.json")
sink = ApiSink("http://localhost:8000/api/v1")

# Create logger
logger = RunLogger(pricing=pricing, sink=sink, default_tags=["production"])

# Instrument your LM
lm = dspy.LM(model="gpt-4o-mini")
dspy.settings.configure(lm=logger.instrument_lm(lm))

# Track runs
with logger.run(tags=["experiment-1"], metadata={"dataset": "qa"}):
    result = await logger.react(my_module, question="What is 2+2?")

Features

• Track LLM token usage and costs
• Support for multiple sinks (API, SQLite, Multi)
• Tool call tracking with @track_tool decorator
• DSPy module integration (ReAct, Predict, ChainOfThought)

License

MIT