Language model
MiMo-V2-Flash
Xiaomi
- Released
- December 16, 2025
- Data date
- September 26, 2026
MiMo-V2-Flash is Xiaomi’s text-only sparse MoE base with about 309B parameters, of which 15B are active per token. The openly available model supports up to 262,144 context tokens and up to 65,536 output tokens. Image and video input are explicitly outside this checkpoint.
Xiaomi describes this model’s hybrid architecture as a foundation for MiMo-V2.5. MiMo-V2-Flash is the text-oriented base, while MiMo-V2.5 is the openly available multimodal agent model.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | APISource |
| Parameters | 309 billionSource |
| Active parameters | 15 billionSource |
| Context window | 262,144 tokensSource |
| Maximum output | 65,536 tokensSource |
| Knowledge Cutoff | December 2024Source |
| License | MITSource |
| Inputs | TextSource |
| Outputs | TextSource |
| Reasoning | YesSource |
| Tool calling | YesSource |
Popularity
36.5 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | models.dev catalog (retrieved September 26, 2026) · Models.dev model catalog · Editorial description reviewed October 2, 2026 |
| Additional source | Hugging Face · MiMo-V2-Flash model card · Editorial description reviewed October 2, 2026 |
| Additional source | Xiaomi MiMo V2.5 model card · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |