MiniMax launches M3.1 Flash, saying it runs AI agent work 70% cheaper and nearly three times faster
Chinese AI company MiniMax released M3.1 Flash, an update built for AI agents, coding and computer use (operating software the way a person would). It says agent tasks cost 70% less to run than on its previous model and that it responds nearly three times faster. Pricing is $0.15 per million tokens of input and $0.60 per million of output (a token is roughly three-quarters of a word). It's available now through MiniMax's API, subscription plan and desktop coding app, and MiniMax says it will release the model for download later, with a more capable Pro version also coming.
What it means for manufacturers and fabricators
Agents that work all day, like chasing quotes or keying in POs, make many model calls per job, so a price drop like this makes it cheaper to hand more busy work to AI. It's another sign that the cost of running agents keeps falling fast.
Related guide: How to Automate Your Business with AI: A Step-by-Step Guide for Manufacturers