

I went with quantized Gemma
Well, was it quantized in a way that iphone 16 supports?
Often it’s the quantization where things break down, and the hardware needs to support the quantization, can’t run FP16 on int8 hardware… And sometimes the act of quantization can cause problems too.
And yeah, LLMs are likely going to be very hit or miss anyway.



Ublock origin has been working for YouTube for me, they’ve had some rough patches where YouTube updated and then they had to update, but it’s been smooth for the past couple years.