OpenAI continues naming chaos despite CEO acknowledging the habit
1 min read
Summary
OpenAI has released GPT-4.1, a new series of language models that outperform GPT-4 in several areas.
The new models will only be available to developers via API, not to regular users of ChatGPT.
This is because the new models have a 1-million-token context window, allowing them to process 3,000 pages of text in one conversation, which competes with Google’s Gemini models.
The GPT-4.5 preview model will be retired for API use, but still accessible through ChatGPT.
OpenAI aims to consolidate its offerings in the future, to simplify its rather confusing model names and versions.
It’s not clear how the new GPT-4.1 versions fit into this plan.
