A survey of over 200,000 software developers has concluded, and the vast majority (over 99%) believe A.I. context size matters. And is, in fact, the most important aspect of any A.I. tooling. Bigger 8===> Better.
New models are released each week by, well, everyone. The new models either have more context, are tuned for specific scenarios, or get internal updates. Every developer has their own preferences when choosing which model to use. Your choice can mean the difference between finishing work early and going to the bar at 2 PM on a Wednesday, or deploying a bug-ridden release that wakes up the boss's boss at 2 AM. And you don't want to anger the boss's boss after they've been drinking since 2 PM on a Wednesday.
Developer Jackson Reaches told us, "When a new model comes out, I look at the context size first, second, and last. If it's bigger than what I'm using now, I switch and never look back. There might be some quirks, but it's a guaranteed net gain. My buddy Hugh Johnson is still using some puny "1M" window. Can you believe it? He just admits it!"
We tracked down the 1% of developers that did not agree context size is king. Of those 2,000 people, 1,905 told us they prefer things along the lines of 'good reasoning', 'accurate results', and 'cheap'.
The remaining 50, all of the women who took part in the survey, agreed size doesn't matter at all. To them, the most important aspect is how you use it. Taking your time at the beginning to set up the prompt goes a long way toward satisfactory results. Keeping engaged with proper stamina, so you can go until the task is properly finished. And of course, a firm backend...API.