Last week, news filtered out that ChatGPT-4, the latest iteration of OpenAI’s chatbot, would be announced, and indeed today is that day. In a Tweet and a new web page, OpenAI has pulled back the covers on its “most capable and aligned model yet:”
here is GPT-4, our most capable and aligned model yet. it is available today in our API (with a waitlist) and in ChatGPT+.https://t.co/2ZFC36xqAJ
it is still flawed, still limited, and it still seems more impressive on first use than it does after you spend more time with it.
— Sam Altman (@sama) March 14, 2023
The webpage links to a fairly technical paper outlining GPT-4, saying:
“…GPT-4, a large-scale, multimodal model which can accept image and text inputs and produce text outputs. While less capable than humans in many real-world scenarios, GPT-4 exhibits human-level performance on various professional and academic benchmarks, including passing a simulated bar exam with a score around the top 10% of test takers.”
This new ChatGPT-4 model is “multi-modal,” meaning it can accept text or image inputs, opening up lots of new possibilities. Microsoft is using OpenAI and a modified version of ChatGPT called the “Prometheus Model” in the new Bing, although it’s not thought to be GPT-4 based, so we’ll have to see how rapidly Bing iterates on this new model.
There will be a webcast at 1pm PDT today on YouTube with more information, we’ll keep you up to date on all the latest.