OpenAI Forks GPT-4 into Separate ChatGPT and API Models

OpenAI has split their GPT-4 model into two distinct versions: one optimized for ChatGPT and another for API use. This change was announced on Twitter, where OpenAI stated they now offer chatgpt-4o-latest in the API while recommending gpt-4o-2024-08-06 for most API usage.

Model Versions

According to OpenAI's documentation, several versions of the GPT-4o model now exist:

• The flagship model, gpt-4o, currently points to gpt-4o-2024-05-13
• gpt-4o-2024-08-06 is the latest snapshot supporting Structured Outputs
• chatgpt-4o-latest is a dynamic model updated to match the current ChatGPT version

Developer Implications

For developers, this fork has practical implications. OpenAI recommends using gpt-4o-2024-08-06 for most API applications, particularly those requiring structured outputs. The chatgpt-4o-latest endpoint suits chat-based applications or when access to the latest conversational features is needed. The chatgpt-4o-latest endpoint may change without notice, while the dated versions offer stability.

What I'm building

Delegate tasks. Get software.

Give Vroni a GitHub issue, bug report, spec, or rough idea. It reads the repo, plans the change, writes code, runs checks, and works toward a review-ready pull request.

Take a look at vroni.com