Hugging Face

Hugging Face

Improve the visibility of your Hugging Face usage with New Relic Hugging Face quickstart.

What's included?

Hugging Face quickstart contains 1 dashboard. These interactive visualizations let you easily explore your data, understand context, and resolve problems faster.
Hugging Face observability quickstart contains 3 alerts. These alerts detect changes in key performance metrics. Integrate these alerts with your favorite tools (like Slack, PagerDuty, etc.) and New Relic will let you know when something needs your attention.
Requests per model
This alert is triggered if requests per model exceeds 1000 in 5 minutes.
Response time
This alert is triggered if response time exceeds 5 seconds in 5 minutes.
This alert is triggered if the number of errors exceeds 2 within 5 minutes.

Why should you monitor your usage of Hugging Face?

Monitor your application powered by Hugging Face language models to ensure, get visibility to what you send to Hugging Face, responses received from Hugging Face, latency, usage and errors. By monitoring the usage, you can infer the cost.

Track the LLM's performance:

Monitor the input & output, latency and errors of your LLM provider. Track performance changes with the providers and versions of your LLM. Monitor usage to understand the cost, rate limits, and general performance.

Track your app:

By tracking key metrics like latency, throughput, error rates, and input & output, you can gain insights into your LangChain app's performance and identify areas of improvement.

Early issue detection:

Detect and address issues early to prevent them from affecting model performance.

Comprehensive Hugging Face monitoring quickstart

Our Hugging Face quickstart provides metrics including error rate, input & output, latency, queries, and lets you integrate with different language models.

What’s included in the Hugging Face quickstart?

New Relic Hugging Face monitoring quickstart provides a variety of pre-built dashboards, which will help you gain insights into the health and performance of your Hugging Face usage. These reports include:

  • Dashboards (average tokens, LLM completion’s details, chain’s details, tool details, top tool names and many more)
  • Alerts (errors, request per model and response time)

How to use this quickstart

  • Sign Up for a free New Relic account or Log In to your existing account.
  • Click the install button.
  • Install the quickstart to get started or improve how you monitor your environment. They're filled with pre-built resources like dashboards, instrumentation, and alerts.
New Relic
Need help? Visit our Support Center or check out our community forum, the Explorers Hub.