Xiaomi retires MiMo-V2.5 models, moves API to multimodal MiMo-V2.6
Xiaomi shut down MiMo-V2.5-Pro and MiMo-V2.5 on October 21, routing API requests to MiMo-V2.6-Pro and MiMo-V2.6-Flash. The switch adds multimodal understanding and raises the Flash model's default completion limit from 32,768 to 131,072 tokens at unchanged API pricing.
- MiMo-V2.5-Pro and MiMo-V2.5 were retired on October 21 at 10 a.m. Beijing time
- Requests to the old model names were rerouted to V2.6 since October 14
- MiMo-V2.6-Flash default completion limit rose from 32,768 to 131,072 tokens
- V2.6 series keeps the previous generation's API pricing
Read next
AI