Model reference · open weights

MiMo-Flash

Available as managed deployment LLMs XiaomiMiMo Text gen 1 variants 93k dl/mo

MiMo-Flash is an open-weight language model from XiaomiMiMo. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released byXiaomiMiMo
TypeLanguage models
TaskText gen
Parameters (lead)309.8B
Context256k tokens
Runs withtransformers
Released2025-12-16
Popularity93k downloads / month
LicenceOpen weights

About

What MiMo-Flash is

|  |  |  |  

MiMo-V2-Flash

MiMo-V2-Flash is a Mixture-of-Experts (MoE) language model with 309B total parameters and 15B active parameters. Designed for high-speed reasoning and agentic workflows, it utilizes a novel hybrid attention architecture and Multi-Token Prediction (MTP) to achieve state-of-the-art performance while significantly reducing inference costs.


1. Introduction

MiMo-V2-Flash creates a new balance between long-context modeling capability and inference efficiency. Key features include:

Read the full model card
  • Hybrid Attention Architecture: Interleaves Sliding Window Attention (SWA) and Global Attention (GA) with a 5:1 ratio and an aggressive 128-token window. This reduces KV-cache storage by nearly 6x while maintaining long-context performance via learnable attention sink bias.
  • Multi-Token Prediction (MTP): Equipped with a lightweight MTP module (0.33B params/block) using dense FFNs. This triples output speed during inference and will be good to accelerates rollout in RL training.
  • Efficient Pre-Training: Trained on 27T tokens using FP8 mixed precision and native 32k seq length. The context window supports up to 256k length.
  • Agentic Capabilities: Post-training utilizes Multi-Teacher On-Policy Distillation (MOPD) and large-scale agentic RL, achieving superior performance on SWE-Bench and complex reasoning tasks.

2. Model Downloads

ModelTotal ParamsActive ParamsContext LengthDownload
MiMo-V2-Flash-Base309B15B256k🤗 HuggingFace
MiMo-V2-Flash309B15B256k🤗 HuggingFace

[!IMPORTANT] We also open-source the 3-layer MTP weights to foster community research.


3. Evaluation Results

Base Model Evaluation

MiMo-V2-Flash-Base demonstrates strong performance across standard benchmarks, surpassing models with significantly larger parameter counts.

CategoryBenchmarkSetting/LengthMiMo-V2-Flash BaseKimi-K2 BaseDeepSeek-V3.1 BaseDeepSeek-V3.2 Exp Base
Params#Activated / #Total-15B / 309B32B / 1043B37B / 671B37B / 671B
GeneralBBH3-shot88.588.788.288.7
MMLU5-shot86.787.887.487.8
MMLU-Redux5-shot90.690.290.090.4
MMLU-Pro5-shot73.269.258.862.1
DROP3-shot84.783.686.386.6
ARC-Challenge25-shot95.996.295.695.5
HellaSwag10-shot88.594.689.289.4
WinoGrande5-shot83.885.385.985.6
TriviaQA5-shot80.385.183.583.9
GPQA-Diamond5-shot55.148.151.052.0
SuperGPQA5-shot41.144.742.343.6
SimpleQA5-shot20.635.326.327.0
MathGSM8K8-shot92.392.191.491.1
MATH4-shot71.070.262.662.5
AIME 24&252-shot35.331.621.624.8
CodeHumanEval+1-shot70.784.864.667.7
MBPP+3-shot71.473.872.269.8
CRUXEval-I1-shot67.574.062.163.9
CRUXEval-O1-shot79.183.576.474.9
MultiPL-E HumanEval0-shot59.560.545.945.7
MultiPL-E MBPP0-shot56.758.852.550.6
BigCodeBench0-shot70.1

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys mimo-flash for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (mimo-flash below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"mimo-flash","messages":[{"role":"user","content":"Hello"}]}'

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms