Project AURIONAI Use-Cases Research & Innovation
Xiaomi
multimodal LLM

MiMo-V2.6-Pro

Xiaomi open model; highest-scoring open-weights model on AA.

XiaomiXiaomiChina
Figure 1: MiMo-V2.6 architecture , omni encoders, hybrid SWA backbone, and MTP blocks.
Levelmodel
Classmultimodal LLM
Parameters1.02T
Context1,000,000
ReleasedSep 2026
Opennessopen weights permissive
Runs onmulti-GPU node
Statusactive

From the model's own page; who made each picture and where it comes from are listed on the Image Credits page.

Benchmark Results

vendor-reportedindependent
Arena Text leaderboardhuman preference arena
independent
1,480 ±9
1,2001,4001,600

Score in Elo; the axis starts at 1,200, as Elo has no zero. Higher is better.

rank 26, Arena Text overall8 Oct 2026 arena.ai
Artificial Analysis Intelligence Index v4.3.2composite index
independent
46
03060

Score in index. Higher is better.

rank 25 of 205, AA leaderboard, 205 current models10 Oct 2026 artificialanalysis.ai
independent
46
03060

Score in index. Higher is better.

rank 1, Artificial Analysis open-weights viewartificialanalysis.ai, artificialanalysis.ai
DeepSWE (version not stated)coding
vendor-reported
71.9
050100

Score in %. Higher is better.

huggingface.co, huggingface.co
Terminal-Bench 2.1agents and tool use
vendor-reported
89.9
050100

Score in %. Higher is better.

huggingface.co, huggingface.co
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

successordistillfinetune
MiMo-V2.5-Pro to MiMo-V2.6-Pro: successor, medium confidenceMiMo-V2.6-Pro to MiMo-V2.6-Pro distilled Qwen-9B variant: distill, medium confidenceMiMo-V2.6-Pro to MiMo-V2.6-Pro-MOPD: finetune, medium confidenceMiMo-V2.6-Pro to MiMo-V2.6-Pro-RL: finetune, medium confidenceMiMo-V2.5-ProMiMo-V2.5-ProXiaomi, Apr 2026MiMo-V2.6-ProMiMo-V2.6-ProXiaomi, Sep 2026MiMo-V2.6-Pro distilled Qwen-9B variantMiMo-V2.6-Pro distilled Qw…XiaomiMiMo-V2.6-Pro-MOPDMiMo-V2.6-Pro-MOPDXiaomiMiMo-V2.6-Pro-RLMiMo-V2.6-Pro-RLXiaomi
XiaomiMiMo-V2.5-Pro is the parent (successor)medium confidencenotes
XiaomiMiMo-V2.6-Pro distilled Qwen-9B variant is a child (distill)medium confidencenotes
XiaomiMiMo-V2.6-Pro-MOPD is a child (finetune)medium confidencenotes
XiaomiMiMo-V2.6-Pro-RL is a child (finetune)medium confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 5 models. Confidence and evidence are the notes' own.

R&D Topics

Efficiency architectures (MoE, sparse attention, long context)produced byOpen-weight frontier releases as research substrateused by

Identity

Developer
Xiaomi
Family
not in the notes
Released
Sep 2026 sourcesregistry dateunverified
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
MoE Sparse MoE (384 experts, 8 active, no shared experts), 60 SWA + 10 global layers, 5-layer MTP
Total parameters
1.02T sources
Active parameters
42B
Checkpoint parameters
–
Context length
1,000,000 tokens
Input
text, image, video, audio
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
MIT License sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
multi-GPU node
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Industry Use Cases

No industry use case links this model.

Sources

  1. LLM Leaderboard - Comparison of AI models from OpenAI, Anthropic, Google, SpaceXAI & others | Artificial Analysis artificialanalysis.ai/leaderboards/models https://artificialanalysis.ai/leaderboards/models
  2. Text Arena: LLM Rankings & AI Model Benchmarks arena.ai https://arena.ai/leaderboard/chat/text
  3. Comparison of Open Source AI Models across Intelligence, Performance, Price, Context Window, and more | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/open-source
  4. LLM Leaderboard - Comparison of AI models from OpenAI, Anthropic, Google, SpaceXAI & others | Artificial Analysis artificialanalysis.ai/leaderboards/models?open_weights=true https://artificialanalysis.ai/leaderboards/models?open_weights=true
  5. huggingface.co/XiaomiMiMo https://huggingface.co/XiaomiMiMo
  6. huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "MiMo-V2.6-Pro", https://aurion.semai.id/models/xiaomi-mimo-v2-6-pro/, records of October 2026.