MiniMax-M3 debuts, beating GPT-5.5 and Gemini 3.1 Pro at 5-10% cost
Key Questions
How does MiniMax-M3 compare to GPT-5.5 and Gemini 3.1 Pro?
MiniMax-M3 outperforms both models on key benchmarks while costing only 5-10% as much, thanks to its sparse attention efficiency technique.
What makes MiniMax-M3 a cost-performance disruptor?
Its dramatically lower pricing combined with superior benchmark results positions it as a serious alternative for high-volume AI workloads.
Is MiniMax-M3 leading in specific evaluations?
It has become the top open model on Next.js agent evaluations according to recent developer reports.
MiniMax-M3 is a new model that outperforms GPT-5.5 and Gemini 3.1 Pro on key benchmarks while costing only 5-10% as much. The sparse attention technique is a clever efficiency hack. This is a serious cost-performance disruptor that my audience needs to know about. No tutorial yet, but strong informational value.
Sources (2)
Updated Jun 2, 2026