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

MiniMax launched M3.1 Flash on October 19th, a model the company says delivers online inference nearly three times faster and costs 70% less to use than its predecessor, according to MiniMax's own blog post. Founder Yan Junjie is positioning the cheaper, faster inference to make AI agent and coding workloads practical to run at scale. MiniMax's speed, price and benchmark figures are company-reported, and the exact baseline behind the reductions has not been independently verified.

MiniMax says M3.1 Flash makes online inference nearly three times faster and cuts costs 70% Founder Yan Junjie is pushing the model into coding and agent workflows; MiniMax's speed, price and benchmark figures are company-reported. By RuntimeWire Staff https://runtimewire.com/author/runtimewire-staff · Published · Updated Primary source: MiniMax https://www.minimax.io/blog/minimax-m31f Why it matters MiniMax is selling inference economics as a way to make agent workloads viable at scale. Its claims are specific, but independent comparisons and the exact baseline behind the speed and cost reductions will determine how much weight customers should give them. MiniMax founder Yan Junjie https://ir.minimax.cn/en/corporate-information/management?ref=runtimewire is betting that cheaper, faster inference can make AI agents more practical to run: on October 19th, MiniMax launched M3.1 Flash https://www.minimax.io/blog/minimax-m31f?ref=runtimewire , a model it says delivers online inference nearly three times faster and costs 70% less to use than its predecessor.… When you purchase through links in our articles, we may earn a small commission https://runtimewire.com/affiliate-monetization-standards . This doesn’t affect our editorial independence.