Skip to content

README.md

Latest commit

 

History

History
64 lines (49 loc) · 2.07 KB

README.md

File metadata and controls

64 lines (49 loc) · 2.07 KB

Gentrace Python SDK Examples

This directory contains simplified examples demonstrating key features of the Gentrace Python SDK.

Examples Overview

Basic Initialization

  • init_simple.py - Basic Gentrace initialization with default OpenTelemetry setup
  • init_manual_otel.py - Manual OpenTelemetry configuration when you need custom control
  • init_with_instrumentation.py - Adding automatic HTTP instrumentation

LLM Integrations

  • openai_simple.py - Simple OpenAI integration with Gentrace tracing
  • openai_instrumentation.py - OpenAI with automatic OpenInference instrumentation
  • openai_agents_instrumentation.py - OpenAI Agents (Swarm) with OpenInference instrumentation
  • anthropic_simple.py - Anthropic Claude integration example
  • pydantic_ai_simple.py - Simple Pydantic AI framework integration

Evaluation

  • eval_simple.py - Basic evaluation example using @eval decorator

Running Examples

1. Install dependencies:

# using rye (recommended)
rye sync --all-features

# using uv
cd ..
uv venv
source .venv/bin/activate
uv pip install ".[openai,openai-agents,anthropic,pydantic-ai]" -r requirements-dev.lock
cd examples

2. Set environment variables:

Option A: Use the provided .env file (recommended)

  • Copy .env.example to .env
  • Update the values with your API keys and configuration

Option B: Set environment variables manually

export GENTRACE_API_KEY="your-api-key"
export GENTRACE_BASE_URL="https://gentrace.ai/api"  # or http://localhost:3000/api
export GENTRACE_PIPELINE_ID="your-pipeline-id"
export OPENAI_API_KEY="your-openai-key"  # For OpenAI examples
export ANTHROPIC_API_KEY="your-anthropic-key"  # For Anthropic examples

3. Run an example:

python init_simple.py

Example Patterns

Each example focuses on one specific use case:

  • Basic init examples show different ways to initialize Gentrace
  • LLM examples demonstrate tracing AI model calls
  • Evaluation examples show how to run experiments and evaluations

For more comprehensive documentation, see the Gentrace docs.