MiMo V2 Flash

Xiaomi's high-efficiency inference model with hybrid architecture, 3 MTP layers for 2.5-3.7x faster inference, and 256K context.

Select Provider