Referencing your other comment, this claim really starts to fall apart.
* You're cherry-picking 2018Q3. So it's not exactly 2 years, it's actually 1.75 years (2018Q3 - 2020Q2).
* But now you're saying pre-covid too, so now it's actually 1.25 years (2018Q3 - 2019Q4).
That period in question is the time after Tesla finished ramping Model 3 production (using a tent!) at Fremont in 2018Q3, and before they finished building the factory in Shanghai in 2020Q1.
So... doesn't it seem reasonable that production gains would be a bit "lumpy"? They go up every time a new factory is finished, and they stay flat until the next one.