PANews reported on March 19th that Xiaomi has released its large-scale agent model, Xiaomi MiMo-V2-Pro. The Xiaomi MiMo-V2-Pro is specifically designed for high-intensity agent workloads in the real world. It boasts over 1TB of total parameters (42B of activation parameters), employs an innovative hybrid attention architecture, and supports ultra-long context lengths of up to 1MB. The MiMo-V2-Pro model is now officially available via API, supporting 1MB context lengths and offering tiered pricing based on usage: Up to 256KB context: $1 input/million tokens, $3 output/million tokens; Up to 1MB context: $2 input/million tokens, $6 output/million tokens. MiMo-V2-Pro is now available on multiple platforms.


