Expanding the GenAI Platform: Now supporting OpenAI models

Posted: April 11, 2025•2 min read

We’re excited to announce a major update to DigitalOcean’s GenAI Platform: support for OpenAI models is now live!

You can now bring your own OpenAI API key to use models like GPT-4o, o1, and o3-mini with GenAI agents on DigitalOcean. Whether you’re testing ideas in the model playground or deploying an agent into production, this gives you more flexibility when choosing the right model for your use case.

Bring your own OpenAI API key to DigitalOcean

GenAI Platform now supports OpenAI’s industry-leading models, allowing you to bring your own OpenAI API key to integrate GPT-4o and reasoning models like o1 and o3-mini into your GenAI agents. Users choose OpenAI’s models for their quality and comparatively advanced capabilities, making them a strong option for a wide range of AI applications.

OpenAI models available:

GPT-4o is a generalist model, designed for conversational AI, content generation, and code assistance.
GPT-4o-mini is designed for fast, efficient task completion across chat, content, and code, with lower latency and cost.
o3-mini is a lightweight reasoning model that excels in root cause analysis, due diligence, debugging, and troubleshooting.

By supporting these models, DigitalOcean provides more flexibility to build AI-powered applications using OpenAI’s technology while seamlessly integrating into our GenAI Platform.

Why use OpenAI models on DigitalOcean?

OpenAI on DigitalOcean’s GenAI Platform gives developers a powerful, streamlined path to production-ready AI, no infrastructure headaches, no re-architecting. OpenAI is one of the most popular names in AI today, with models known for their performance, reliability, and capabilities. By integrating OpenAI models with DigitalOcean’s GenAI Platform, you can:

Use OpenAI’s powerful models without setting up your own infrastructure, it’s all handled by us
Deploy AI agents with just a few clicks, using models from OpenAI and built-in tools like function calling and guardrails from DigitalOcean
Connect your own data to OpenAI models easily, so your AI can answer questions and take actions specific to your business
Get predictable pricing and easy monitoring, all through the same platform you use to run the rest of your app

Try before you build in the model playground

Don’t have a key yet? No problem. You can now try out OpenAI models in the model playground, no API key required. Get a feel for how each model performs across your use cases before committing to them in production. This is a great way to compare model behavior and find the right fit for your AI agent.