Matt Wolfe picks GPT-5.5 as his favorite LLM right now for its range, and the GPT image model for text-heavy visuals. He switches based on what works best, not loyalty. A month ago he preferred Claude Opus 4.7 for coding.
If you had to choose only one LLM, ChatGPT, Gemini, Perplexity, Claude, Copilot, Grok, or ZAI, which one are you rolling with and why? Same for image and video models. You can't say Leonardo, no cheating. Okay, so when it comes to this question, I have zero, zero loyalty. Like I will use ChatGPT when I feel like it is the best model of the moment. I will go and use the Claude models if I feel like they're the best models of the moment. As of this recording, GPT-5.5 is my go-to, But if you would've asked me last month, it was probably Claude Opus 4.7. I just use whatever I feel is the best model in the moment. At one point, I really liked the conversation abilities of ChatGPT, but the coding abilities of Claude, so I would jump between them based on what I was using it for. Right now, I kind of find GPT-5.5 to be good at almost everything, so I'm primarily using that. Will my opinion change next week or next month or 3 months from now? Absolutely. It will probably keep changing as new models come out. Same for image and video models. Models. If you would have asked me a month ago, it was Nano Banana Pro. That model was the best of the best for me. You ask me today, I really, really like the GPT image model, especially if I want text in it. Again, zero loyalty.
AI News: Dots, GPT-6.1 Sol, Sonnet 5.5, Gemini 4, and everything you need to know
