Xiaomi MiMo-V2-Flash is a Xiaomi model built for long-context use, hybrid attention, and Multi-Token Prediction in the MiMo series.
Capabilities, design details, and architectural traits
MiMo-V2-Flash is the series model behind the MiMo-V2.5 design. It is tied to the same long-context architecture and the same hybrid attention and Multi-Token Prediction pattern.
| Differentiator | Documented trait |
|---|---|
| Hybrid attention | Uses hybrid attention. |
| Multi-Token Prediction | Uses Multi-Token Prediction, shortened as MTP. |
| Long-context design | Serves as the base for the MiMo-V2.5 context and trajectory design. |
The model is defined by its architecture rather than a chat-style role. The documented identity is its use in the MiMo series as the source of the hybrid attention and MTP setup.
MiMo-V2-Flash sits inside Xiaomi MiMo’s model family and connects directly to the V2.5 series. That makes it a structural model in the lineup, not just a named endpoint.
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Xiaomi | Anthropic | Anthropic |
| Release Date | December 16, 2025 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | - | May 2026 | - |
| Context & Limits | |||
| Context Window | 262K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.10 Best Input Pricing | $5 | $10 |
| Output Pricing | $0.30 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | text | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 25.1 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | 49.8 | 78.0 Best Coding Index | 76.5 |
| Agentic Index | 12.0 | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
