Even if Kimi is better on BenchMarks, MiniMax seems to be a lot better in backend code in the wild. Our codebase is well over 500k lines of code and MiniMax seems to navigate it a lot better than Kimi.
And you can clearly see that on the statistics of Kilo, that people like MiniMax’s model still a lot better than Kimi’s model:
MiniMax also just wins on price/performance. For example this is the pricing of the two models at Fireworks:
Guess which one is Kimi and which one is MiniMax :P
A thought from Daniel.
Thanks for spending a little time here.
Thanks for spending a little time here.