Systems Engineer in public safety focused on tech resilience. I write about local AI, virtualization, and cybersecurity. Currently building secure, sovereign home labs and documenting what I learn.
At this point, the surrounding UX matters more to me than the raw model itself.
ChatGPT Projects.
Actually usable TTS/STT.
Perplexity feeling strangely human because of the ElevenLabs-style voice experience.
Grok being willing to discuss things other models instantly refuse.
Meanwhile some models are technically "smarter", but then the mobile app keeps the screen on during voice conversations and the whole illusion collapses. Feels like we're no longer comparing just LLMs. We're comparing entire ecosystems around them. And overall, ChatGPT still feels the most complete to me.
Systems Engineer in public safety focused on tech resilience. I write about local AI, virtualization, and cybersecurity. Currently building secure, sovereign home labs and documenting what I learn.
I use got and Claude mostly and Claude more since I get more in depth accurate responses. For me accuracy and natural language prompts are more important.
I use my own UI but Claude projects allow a lot more files to be added when you are creating something complex that has a lot of data.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
Top comments (3)
I like lImagen 2 for naturula photos, gemmini for seacrh and naturaul conversational use, claude is my go to for most everything else.
At this point, the surrounding UX matters more to me than the raw model itself.
ChatGPT Projects.
Actually usable TTS/STT.
Perplexity feeling strangely human because of the ElevenLabs-style voice experience.
Grok being willing to discuss things other models instantly refuse.
Meanwhile some models are technically "smarter", but then the mobile app keeps the screen on during voice conversations and the whole illusion collapses. Feels like we're no longer comparing just LLMs. We're comparing entire ecosystems around them. And overall, ChatGPT still feels the most complete to me.
I use got and Claude mostly and Claude more since I get more in depth accurate responses. For me accuracy and natural language prompts are more important.
I use my own UI but Claude projects allow a lot more files to be added when you are creating something complex that has a lot of data.