[MikeB]

GPT-4o (“o” for “omni”) is a step towards much more natural human-computer interaction—it accepts as input any combination of text, audio, and image and generates any combination of text, audio, and image outputs. It can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time(opens in a new window) in a conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models.

https://openai.com/index/hello-gpt-4o/

[MikeB]

They are actually claiming reduced tokens in language. "English 1.1x fewer tokens (from 27 to 24)", with some languages compressing more.

Although there's little to say how it works, as Noam Chomsky points out, nothing is learned from language if a computer arbitrarily uses language and maths to predict with, including arbitrary pre-compression...

Response times are good however.

ChatGPT 3.5 is fast as well. https://chatgpt.com/