Online LLM Evaluations

Arize now lets you run automated actions on your LLM spans, including evaluations. With online evaluations, ongoing evals run automatically as new spans come into Arize from your LLM app. For development, you can automatically run an evaluation on every trace that doesn't have an evaluation yet. In production, you can sample a set of your traffic to run evaluations for monitoring. Evaluation scripts run every five minutes.In this quick demo – part III in our series on LLM evals – Eric Xiao shows you how to quickly set up online LLM evals in the Arize platform, including how to select your traces, select your LLM evaluator, and then create your evaluation template.
Additional Resources
🔗 How to setup tasks to run ongoing evaluations
🔗 LLM Evaluation
Recommended resources
Prompt Playground
A prompt playground offers a UI to experiment with prompt templates, input variables, LLM models and LLM parameters.…
Read more
Tracing LLM Function Calls in Arize
Learn how to simplify the debugging process by logging chat history and function calls with a single line…
Read more
AB Testing for LLM Applications
Don’t ship vibes.
Arize gives AI teams observability and evals to understand and improve agent performance.