Xiaomi

Provider of advanced AI models and technologies.

1 Featured Models

1 Featured Text Models

Compare Models

Key Highlights

MiMo models emphasize long-context reasoning and agent scenarios.

Mixture-of-Experts architecture for strong performance per active parameter.

Hybrid attention designed to scale to very long context windows.

Open-source releases aimed at broad community adoption.

MiMo-V2-Flash

Dec 2025

MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting hybrid attention architecture. MiMo-V2-Flash supports a hybrid-thinking toggle and a 256K context window, and excels at reasoning, coding, and agent scenarios. On SWE-bench Verified and SWE-bench Multilingual, MiMo-V2-Flash ranks as the top #1 open-source model globally, delivering performance comparable to Claude Sonnet 4.5 while costing only about 3.5% as much.

conversationreasoningcode-generationanalysisagentic-tool-use