Xiaomi, released 2026-09-21. Sparse MoE decoder, 309B total, 15B active, 1M context. Attention: GQA with sliding-window/global attention.
Sparse MoEXiaomi 🇨🇳 · December 16, 2025
Xiaomi MiMo-V2-Flash (309B)
Large MoE model that pushes sliding-window attention harder than most contemporaries.
Key detail
Uses an unusually small 128-token local window plus multi-token prediction.Total params
309B
Active / token
15B
Context
256K
Fact sheet
- Scale
- 309B total, 15B active (4.9% active)
- Context window
- 262,144 tokens
- Decoder type
- Sparse MoE
- Attention
- 5:1 sliding-window/global attention
- Layer mix
- 40 sliding-window + 8 global
- KV cache / token (bf16)
- 144 KiB · Moderate
- License
- MIT License
- Released
Efficiency
Parameters active per token
4.9%
15B of 309B weights run for each token.
AA Intelligence Index v4.3.2
Source16
- Scientific
- 4.3
Architecture concepts
Sources & resources
Use the ← and → keys to step through releases.
Similar architectures
Xiaomi, released 2026-04-22. Sparse MoE decoder, 310B total, 15B active, 1M context. Attention: 5:1 sliding-window/global attention.
Motif Technologies, released 2026-07-21. Sparse MoE decoder, 314B total, 13B active, 256K context. Attention: GDLA with MLA-style latent projections, RoPE, and 3:1 sliding-window/full attention.
MiniMax, released 2026-02-12. Sparse MoE decoder, 230B total, 10B active, 192K context. Attention: GQA with QK-Norm.