GGUF
I wanted to send Bartowski a private message on here—but I cannot figure out how.
Is GGUF 'more intelligent' than EXL2 quants?
The reason I ask—I was under the impression that GGUF was smarter. But someone told me that EXL2 is smarter and faster than GGUF. I read elsewhere on reddit (from a thread 5 months ago) that some people felt GGUF was smarter due to the addition of iMatrix. However, I normally run Q6 and Q8 GGUF quants...so unsure if iMatrix is a benefit at those quant levels.
The reason I ask this question—I use AI to help me work. I treat the AI as if it were a live human being sitting next to me. I am a copywriter. So, I bounce a lot of ideas off AI and I'm always on the hunt for a model that can bring that 'wow' factor to the table in terms of how intelligent it is and its understanding of the King's and all the nuances. That's what I mean by 'intelligence.'