[stavros]

True, I just tried it for generating a book summary, and Sonnet 3.5 was very bad. GPT-4o is equally bad at that , gpt-4-turbo is great.

[netsec_burn]

This more likely has to do with context length?

[stavros]

No, all the information is there, but gpt-4o tends to produce bullet points (https://www.thesummarist.net/summary/the-making-of-a-manager...), whereas gpt-4-turbo tends to produce much more readable prose (https://www.thesummarist.net/summary/supercommunicators/the-...).

[Obscurity4340]

How is prose more readable than bullets?

[stavros]

* Clearer narrative

* Connection between points

* Flows better

* Eyes don't start-stop as much

[Obscurity4340]

I think I was thinking more along the lines of if I'm looking for specific information or to get a condensed understanding of something

Different readable than the more flowing, conjunct readable than yours (which is the more typical use of it I concede)