How to beginner · 3 min read

DSPy pipeline explained

Q: DSPy pipeline explained

A DSPy pipeline is a declarative way to compose AI tasks by defining Signatures and chaining them with dspy.Predict or dspy.ChainOfThought. It enables structured, reusable AI workflows with simple Python classes and method calls.

Quick answer

A DSPy pipeline is a declarative way to compose AI tasks by defining Signatures and chaining them with dspy.Predict or dspy.ChainOfThought. It enables structured, reusable AI workflows with simple Python classes and method calls.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install dspy openai>=1.0

Setup

Install the dspy package and set your OpenAI API key as an environment variable.

Run pip install dspy openai to install dependencies.
Set your API key in your shell: export OPENAI_API_KEY='your_api_key' (Linux/macOS) or setx OPENAI_API_KEY "your_api_key" (Windows).

bash

pip install dspy openai

Step by step

Define a dspy.Signature class to specify input and output fields, then create a dspy.Predict instance to run the AI model. This example extracts a summary from a given text.

python

import os
import dspy
from openai import OpenAI

# Configure DSPy with OpenAI LLM
client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
llm = dspy.LM("openai/gpt-4o-mini", client=client)
dspy.configure(lm=llm)

# Define a Signature for the pipeline
class Summarize(dspy.Signature):
    text: str = dspy.InputField()
    summary: str = dspy.OutputField()

# Create a Predict instance
summarize = dspy.Predict(Summarize)

# Run the pipeline
result = summarize(text="DSPy is a declarative Python library for AI pipelines.")
print("Summary:", result.summary)

output

Summary: DSPy is a Python library that simplifies building AI pipelines declaratively.

Common variations

You can use dspy.ChainOfThought for step-by-step reasoning pipelines or switch models by changing the dspy.LM initialization. Async usage is not currently supported directly.

python

import dspy

# Chain of Thought example
class QA(dspy.Signature):
    question: str = dspy.InputField()
    answer: str = dspy.OutputField()

qa = dspy.ChainOfThought(QA)
result = qa(question="What is DSPy?")
print("Answer:", result.answer)

output

Answer: DSPy is a declarative Python library that simplifies AI programming by defining structured tasks.

Troubleshooting

If you see API key missing, ensure OPENAI_API_KEY is set in your environment.
If the model call fails, verify your internet connection and API quota.
For unexpected output, check your Signature fields match the prompt requirements.

✅

Key Takeaways

Use dspy.Signature to define structured AI tasks with typed inputs and outputs.
Create pipelines with dspy.Predict for simple calls or dspy.ChainOfThought for reasoning.
Configure DSPy with an OpenAI client using your API key from environment variables.
DSPy pipelines enable clean, reusable, and declarative AI workflows in Python.
Check environment variables and model names if you encounter API errors.

Verified 2026-04 · openai/gpt-4o-mini

Verify ↗