Examples

Explore the capabilities of Phoenix with notebooks

Agent Examples

Example agents are fully instrumented with OpenInference and utilize end-to-end tracing with Phoenix for comprehensive performance analysis.

Code Generation Agent

Explore a Code Generator Copilot Agent designed to generate, optimize, and validate code.

RAG Agent

Enter a source URL and collect traces in Phoenix to see how a RAG Agent can retrieve and generate accurate responses.

Computer Use Agent

Test out a Computer Use (Operator) Agent built to execute commands, edit files, and manage system operations.

Agents Cookbook

Build a customer support agent to trace activity, assess performance, and experiment with prompts and models.

Application Tutorials

Example full-stack applications instrumented using OpenInference and observed via Phoenix server instances.

LLM Applications - Tracing, Evaluation, and Analysis

Build, observe, evaluate, and troubleshoot an LLM-powered chatbot using tracing, LLM Evals, and clustering.

OpenAI

Evaluating a Search & Retrieval Application

Improve a LlamaIndex search and retrieval application by identifying documentation gaps and analyzing retrievals.

OpenAI, LlamaIndex

TypeScript Chatbot with User Feedback

Instrument, monitor, and analyze a TypeScript application with built-in user feedback via the chat interface.

OpenAI, Express, Next.JS

Python Chatbot with User Feedback

Manually instrument a Python chatbot with integrated user feedback, generating trace data for in-depth insights.

OpenAI, Streamlit

LLM Traces

Trace through the execution of your LLM application to understand its internal structure and to troubleshoot issues with retrieval, tool execution, LLM calls, and more.

Tracing Applications

OpenAI Application

LlamaIndex Application

DSPy Application

Haystack Application

Groq Application

CrewAI Application

Tracing Use Cases

LangChain OpenAI Agent

LlamaIndex OpenAI Agent

OpenAI Structured Data Extraction Service

RAG Chatbot Application

LangChain + OpenAI RAG Application

LlamaIndex + OpenAI RAG Application

Tracing with Sessions

OpenAI (Python)

OpenAI (JS/TS)

LlamaIndex

Datasets and Experiments

Iteratively improve your LLM task by building datasets, running experiments, and evaluating performance using code and LLM-as-a-judge.

Comprehensive Use Cases

Text2SQL

Summarization Service

Email Text Extraction

Pairwise Evaluator

RAG Use Cases

Answer and Context Relevancy Evals

Response Guideline Evals

LlamaIndex RAG with Reranker

LLM Evals

Leverage the power of large language models to evaluate your generative model or application for hallucinations, toxicity, relevance of retrieved documents, and more.

Common Evaluations

Hallucination Evals

Toxicity Evals

Summarization Evals

Evaluations Use Cases

Retrieved Document Relevance

Code Readability Evals

Question-Answering Evals

Evaluating and Improving RAG Applications

End-to-End RAG Application Evaluation

LlamaIndex Application

LlamaIndex Application using Milvus Vector Store

Data Analysis: Embeddings & Structured Data

Use embeddings to explore lower-dimensional representations of your data, identifying clusters of high drift and performance degradation. Complement this with statistical analysis of structured data for A/B testing, temporal drift detection, and deeper performance insights.

Embeddings Analysis: Data Exploration

Analyze GPT Prompt and Response Data

Find Clusters, Export, and Explore with GPT

Embeddings Analysis: Model Performance

Active Learning for a Drifting Image Classification Model

Root-Cause Analysis for a Drifting Sentiment Classification Model

Troubleshooting an LLM Summarization Task

Structured Data Analysis

Detecting Fraud with Tabular Embeddings