Chinese artificial intelligence startup MiniMax has launched an upgraded version of its M3.1 Flash model, promising faster processing, improved capabilities and lower operating costs as competition in the global AI industry intensifies.
Announced on Monday, October 19, 2026, the updated model is designed to improve performance in complex automated tasks, software programming, computer use and cybersecurity.
MiniMax said the new version represents a major improvement over its predecessor, with online inference speeds nearly tripling and usage costs falling by 70%.
The improvements could make AI-powered tools more accessible to businesses and developers seeking to automate operations, develop software and strengthen cybersecurity without incurring high computing expenses.
The model’s enhanced capabilities also reflect the growing competition among Chinese technology companies to develop AI systems that can compete internationally on performance, speed and affordability.
Related News
- US and Chinese Investors Keep AI Money Flowing Across Rival Supply Chains https://vistaonenews.com/2026/09/22/us-and-chinese-investors-keep-ai-money-flowing-across-rival-supply-chains/
- DeepSeek Taps CITIC Securities for Planned Shanghai IPO https://vistaonenews.com/2026/09/09/deepseek-taps-citic-securities-for-planned-shanghai-ipo/
Agentic AI, which enables systems to perform multistep tasks with limited human intervention, has become an increasingly important area of development as businesses explore ways to automate complex workflows.
MiniMax’s latest release highlights the industry’s focus on making AI models more efficient while expanding their practical applications.
However, the company’s performance and cost figures are its own claims, and independent testing will be important in determining how the updated model compares with competing systems in real-world applications.
Leave a comment