How to beginner · 3 min read

How to configure DSPy LM

Q: How to configure DSPy LM

To configure a dspy.LM, import dspy and instantiate dspy.LM with the model name and your API key from os.environ. Then call dspy.configure(lm=lm) to set it as the default language model for DSPy workflows.

Quick answer

To configure a dspy.LM, import dspy and instantiate dspy.LM with the model name and your API key from os.environ. Then call dspy.configure(lm=lm) to set it as the default language model for DSPy workflows.

PREREQUISITES

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

Setup

Install the dspy package and ensure your OpenAI API key is set in the environment variable OPENAI_API_KEY. This key is required to authenticate requests to the OpenAI API.

bash

pip install dspy openai>=1.0

Step by step

Import dspy and os. Create an instance of dspy.LM with your model and API key from os.environ. Then configure DSPy to use this LM instance. Finally, define a signature and run a prediction.

python

import os
import dspy

# Instantiate the DSPy LM with model and API key
lm = dspy.LM("openai/gpt-4o-mini", api_key=os.environ["OPENAI_API_KEY"])

# Configure DSPy to use this LM
dspy.configure(lm=lm)

# Define a signature for a simple QA task
class QA(dspy.Signature):
    question: str = dspy.InputField()
    answer: str = dspy.OutputField()

# Create a prediction callable
qa = dspy.Predict(QA)

# Run a prediction
result = qa(question="What is DSPy?")
print(result.answer)

output

DSPy is a declarative Python library for building AI applications with clean, type-safe signatures.

Common variations

Use different OpenAI models by changing the model string, e.g., "openai/gpt-4o".
Configure DSPy with other LMs by instantiating dspy.LM with different providers.
For async usage, DSPy supports async signatures and calls.

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY is correctly set in your environment.
If the model is not found, confirm the model name is valid and supported by your API key.
For import errors, ensure dspy and openai packages are installed and up to date.

✅

Key Takeaways

Always instantiate dspy.LM with your API key from os.environ for security.
Call dspy.configure(lm=lm) to set the default LM for DSPy workflows.
You can switch models easily by changing the model string in dspy.LM.
DSPy supports both synchronous and asynchronous usage patterns.
Check environment variables and package versions to resolve common errors.

Verified 2026-04 · openai/gpt-4o-mini

Verify ↗