MiniMax says M3.1 Flash makes online inference nearly three times faster and cuts costs 70%

Summary by forck

MiniMax launched M3.1 Flash on October 19th, a model it says runs online inference nearly three times faster and costs 70% less to use than its predecessor.

• Founder Yan Junjie is betting that cheaper, faster inference can make AI agents more practical to run at scale, with MiniMax selling inference economics as the way to make agent workloads viable.
• The company's claims are specific, but the source notes that independent comparisons and the exact baseline behind the speed and cost reductions will determine how much weight customers should give them.