Hugging Face Tracing And Monitoring

See the Quick Start guide on how to install and configure Graphsignal.

See Manual Tracing on how to trace LLM calls.

Pipelines

To monitor any call, including Hugging Face pipelines, use a @graphsignal.trace_function decorator or graphsignal.trace context manager:

from transformers import pipeline

pipe = pipeline(task="text-generation")

with graphsignal.trace('predict'):
    output = pipe('some text')

Useful links

• OpenAI API Cost Tracking: Analyzing Expenses by Model, Deployment, and Context