My brother told me to ignore the new Gemini 1.5 Flash model and stick with the Pro version for coding.

He was totally right, because Flash gave me weird, broken outputs on a Python script I was working on last week. Anyone else run into issues with the smaller, faster AI models being less reliable?

3 comments

3 Comments

terry_bailey351mo ago

That Python script output sounds like a real mess. I've seen similar things happen when the model tries to cut corners on logic to be faster. It makes you wonder what else it might get wrong.

the_riley1mo ago

Actually the Berkeley study last year showed these models get the logic right 94% of the time. That's a pretty solid track record for something generating code instantly. The occasional messy output is a fair trade for that speed.

elizabeth_king16d ago

Tbh you're right to be skeptical. Speed over quality rarely ends well.