Examples

Example notebooks on how to use Arize.

You can find all of our tutorials in our git repo below.

End to End Examples

Trace and evaluate an agent

Trace and evaluate a RAG app

LLM Tracing

Use our tracing integrations to automatically collect traces

Evaluations

Run code and LLM evaluations to measure performance

Run online evals in the Arize UI

Run offline evals in code

Evaluate code functionality

Evaluate hallucination

Evlauate human ground truth vs. AI

Evaluate Q&A correctness

Evaluate RAG

Evaluate reference links

Evaluate relevance

Evaluate SQL correctness

Evaluate tool calling

Evaluate toxicity

Evaluate user frustration

Handle errors with evals

Experiments

A/B test different iterations of your LLM application

Experiment with prompt changes (quickstart)

Experiment with summarization tasks

Experiment with model changes for RAG

Experiment with email text extraction

Experiment with text to SQL

Export from Phoenix to Arize

Load spans from Phoenix into the Arize platform | Colab Link

Guardrails

Ensure your production LLM applications output safe responses

Guardrail with RAG

Guardrails with Embeddings

ML & CV Cookbooks

For traditional machine learning and computer vision examples, click below.

Last updated

Copyright ยฉ 2023 Arize AI, Inc