Chinese AI company MiniMax releases M3.1 Flash, 70% cheaper and 3x faster
The gist
MiniMax released M3.1 Flash, claiming online inference nearly three times faster and usage costs 70% lower than its predecessor.
An independent evaluator scored it 42 on intelligence but flagged a 93% hallucination rate.
How it unfolded
- Sep 27, 2026MiniMax launched M3.1-Flash-Preview inside its MiniMax Code agent tool with no model card, benchmark report or public pricing, available to Token Plan subscribers.
- Oct 19, 2026MiniMax officially rolled out MiniMax M3.1 Flash, saying online inference is nearly three times faster and usage costs are 70% lower than its predecessor.
- Oct 19, 2026MiniMax disclosed that Space Bunny, which processed more than 200 trillion tokens while operating anonymously, was an early version of M3.1 Flash.