In brief
- MiniMax-M1 excels at coding and agent tasks, but creative writers will want to look elsewhere.
- Despite marketing claims, real-world testing finds platform limits, performance slowdowns, and censorship oddities.
- Benchmark scores and feature set put MiniMax-M1 in direct competition with paid U.S. models—at zero cost.
A new AI model out of China is generating sparks—for what it does well, what it doesn’t, and what it might mean for the balance of global AI power.
MiniMax-M1, released by the Chinese startup of the same name, positions itself as the most capable open-source “reasoning model” to date. Able to handle a million tokens of context, it boasts numbers on par with Google’s closed-source Gemini 2.5 Pro—yet it’s available for free. On paper, that makes it a potential rival to OpenAI’s ChatGPT, Anthropic’s Claude, and other U.S. AI leaders.
Oh yeah—it also beats fellow Chinese startup DeepSeek R1’s capabilities in some respects.
Go to Source to See Full Article
Author: Jose Antonio Lanz
