This week I've been testing different LLM's
Chatgpt/Grok/Gemini/Claude/Perplexity Pro (got a year for free that's why 🤓)
What have I liked using each of them briefly you ask ?
Well, I enjoyed the speaking conversational mode of ChatGPT. You can explore whatever realm of knowledge you wish to while driving for exampe. It has a well rounded output for anything you might throw at him but the other AI models might win at specific queries.
For example, Gemini is excellent with his Nano Banana integrated model to edit photos and to create pictures from scratch just sending a prompt. It makes it very easy and que quality is very impressive. I also liked copy and pasting YT videos so it summarizes them for me.
But I found out that sometimes it does not do that work excusing itself with "I was unable to retrieve the information or transcript for the YouTube video at the URL you provided."
When that happened I just went to Perplexity and I got perplexed receiving the answer in less than 5s. Truly astonishing. This AI has to be used more like a buffed search engine so I'm trying to adapt to be my new Google search bar.
Grok. Haven't used it too much but I loved a lot a health related question that I had which I sent first to chatgpt. But Grok's answer deserved a 10/10 if I was a teacher. Editing photos I did not like it whatsoever. Maybe my prompting has to be refined but for now I'll go first to Gemini and Perplexity for sure.
And lastly, Claude. My least used LLM. Great at coding as I heard and also at writing. Made a couple of queries, its answer seemed very complete and thorough.
This is all for now. I have interest in Automations and getting good at prompting is mandatory, aka writing your instructions comprehensible enough for the LLM's to understand what you want at first try.
DbK, ST.