Space Bunny (MiniMax M3.1 Flash) vs DeepSeek V4.1 Flash on the same 3D piano task: 46 vs 49 minutes
Both models completed the same 3D gummy piano task, with MiniMax at 46 minutes and DeepSeek at 49; the author also compares API pricing, noting MiniMax is flat-rate while DeepSeek doubles at peak.
Space Bunny did 200T+ tokens before anyone knew the name Turns out it was MiniMax M3.1 Flash I gave it and DeepSeek V4.1 Flash the same task: a 3D gummy piano with squishy keys, falling bubbles and a tempo control Both got it working MiniMax finished in 46 min, DeepSeek took 49 API price per 1M tokens: - MiniMax: $0.15 in, $0.60 out, any time - DeepSeek: same off-peak, double at peak hours MiniMax M3.1 Flash did better, what do you think?
This post quotes another post: x.com/minimax_ai/status/2112150867868541221