Back to Xiaomi MiMo-V2.6-Pro-RL 1.02T in the gallery

Xiaomi

Omnimodal sparse MoE model with the MiMo-V2.5-Pro text backbone, mixed-domain reinforcement learning, and a five-layer DFlash speculative decoder.

Xiaomi MiMo-V2.6-Pro-RL 1.02T architecture diagram
Xiaomi MiMo-V2.6-Pro-RL 1.02T architecture diagram. Select the diagram to view it at full size.

Architecture and specifications

Scale
1.02T total, 42B active (4.1% active)
Context (tokens)
1,048,576
Vocabulary
152,576 (~152k)
License
MIT License
Date
2026-09-21
Decoder type
Sparse omnimodal MoE
Attention
GQA with 6:1 sliding-window/global attention
Layer mix
60 sliding-window + 10 global
KV cache / token (bf16) info
350 KiB · Very high
Key detail
Uses 384 routed experts with top-8 routing, one dense prefix layer, no shared experts, and a 128-token sliding window.

Training and benchmarks

Pretraining data
30T tokens 27T text-stage tokens plus 3T omni-stage tokens. Technical report
Post-training methods
SFT, RLVR, Distillation Short SFT, mixed-task GRPO using tests and agentic graders, and MOPD2 distillation described in the series recipe. Technical report
AA Intelligence Index v5.0 info
Total score 29.9 Knowledge Work 45.9 Coding 28.3 Science 17.0 General 28.0
AA Intelligence Index v4.3.2 info
Total score 46.3 General 42.3 Scientific 38.0 Coding 47.9 Agents 54.9

Further reading and tools