Skip to main content
In this guide, we’ll show how Ollama’s seamless compatibility with the OpenAI API lets you build and test LLM-powered applications locally—and then switch to the OpenAI cloud for production with zero code changes. You’ll learn how to configure your environment variables, compare development versus production setups, and follow a real-world workflow.

​
Why Use OpenAI Compatibility?

By leveraging the OpenAI client libraries against a local Ollama endpoint, you get:
  • Consistent API interface across development and production
  • Zero code rewriting when moving to the cloud
  • Full control for local testing without incurring API costs
Let’s follow Jane’s journey from local development to production-ready deployment.
The image illustrates "Jane's Story," showing a progression from "Jane" to "Development" with Ollama, and then to "Production" with OpenAI Library.

​
1. Development Environment Setup

In development, point your OpenAI client at Ollama’s REST API. Add these lines to your .env file: 
# .env (Development)
OPENAI_API_KEY=anyrandomtext
LLM_ENDPOINT="http://localhost:11434/v1"
MODEL=llama3:2:1b
Ollama does not validate OPENAI_API_KEY locally. Feel free to use a placeholder value while testing.
Then initialize your OpenAI client in code as usual: 
import OpenAI from "openai";
const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
  baseURL: process.env.LLM_ENDPOINT,
});

​
2. Production Environment Setup

When you’re ready to go live, sign in to the OpenAI dashboard to create an API key. Update your .env as follows: 
# .env (Production)
OPENAI_API_KEY=sk-XXXXXXXXXXXXXXXXXXXXXXXX
LLM_ENDPOINT="https://api.openai.com/v1"
MODEL=gpt-3.5-turbo
Keep your real OPENAI_API_KEY secure. Never commit it to source control or expose it in client-side code.

​
Configuration Comparison

EnvironmentOPENAI_API_KEYLLM_ENDPOINTMODEL
Developmentanyrandomtexthttp://localhost:11434/v1llama3:2:1b
ProductionYour OpenAI API keyhttps://api.openai.com/v1gpt-3.5-turbo
No changes to your application code are required—just swapping environment variables.

​
3. Next Steps

  1. Generate or rotate your OpenAI API keys via the OpenAI dashboard.
  2. Deploy your application, ensuring the production .env is configured.
The image outlines two next steps: generating API keys in OpenAI and using these keys with an application.

​
References

Watch Video