Its a genuinely nice idea, I do love the second clock face you've done, the sort of "fuel gauge" one. I think a great next step would be some sort of gallery of clock faces that people can use and contribute to, based on whatever code you've used to create them
> This class of bug seems to be in the harness, not in the model itself. It’s somehow labelling internal reasoning messages as coming from the user, which is why the model is so confident that “No, you said that.”
from the article.
I don't think the evidence supports this. It's not mislabelling things, it's fabricating things the user said. That's not part of reasoning.
This is great, and I'm not knocking it, but every time I see these apps it reminds me of my phone.
My 2021 Google Pixel 6, when offline, can transcribe speech to text, and also corrects things contextually. it can make a mistake, and as I continue to speak, it will go back and correct something earlier in the sentence. What tech does Google have shoved in there that predates Whisper and Qwen by five years? And why do we now need a 1Gb of transformers to do it on a more powerful platform?
https://electronics.sony.com/audio/walkman-digital-recorders...
The cheapest walkman model is $399.99