MiMo-Audio-7B-Instruct

<div align="center"> <picture> <source srcset="https://github.com/XiaomiMiMo/MiMo-VL/raw/main/figures/Xiaomi_MiMo_darkmode.png?raw=true" media="(prefers-color-scheme: dark)"> <img src="https://github.com/XiaomiMiMo/MiMo-VL/raw/main/figures/Xiaomi_MiMo.png?raw=true" width="60%" alt="Xiaomi-MiMo" /> </picture> </div> <h3 align="center"> <b> <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span> <br/> MiMo Audio: Audio Language Models are Few-Shot Learners <br/> <span>━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━</span> <br/> </b> </h3> <br/> <div align="center" style="line-height: 1;"> | <a href="https://github.com/XiaomiMiMo/MiMo-Audio" target="_blank">🤖 GitHub</a> &nbsp;| <a href="https://github.com/XiaomiMiMo/MiMo-Audio/blob/main/MiMo-Audio-Technical-Report.pdf" target="_blank">📄 Paper</a> &nbsp;| <a href="https://xiaomimimo.github.io/MiMo-Audio-Demo" target="_blank">📰 Blog</a> &nbsp;| <a href="https://github.com/XiaomiMiMo/MiMo-Audio-Eval" target="_blank">📊 MiMo-Audio-Eval</a> &nbsp;| <br/> </div> <br/>

Introduction

Existing audio language models typically rely on task-specific fine-tuning to accomplish particular audio tasks. In contrast, humans are able to generalize to new audio tasks with only a few examples or simple instructions. GPT-3 has shown that scaling next-token prediction pretraining enables strong generalization capabilities in text, and we believe this paradigm is equally applicable to the audio domain. By scaling MiMo-Audio's pretraining data to over one hundred million of hours, we observe the emergence of few-shot learning capabilities across a diverse set of audio tasks. We develop a systematic evaluation of these capabilities and find that MiMo-Audio-7B-Base achieves SOTA performance on both speech intelligence and audio understanding benchmarks among open-source models. Beyond standard metrics, MiMo-Audio-7B-Base generalizes to tasks absent from its training data, such as voice conversion, style transfer, and speech editing. MiMo-Audio-7B-Base also demonstrates powerful speech continuation capabilities, capable of generating highly realistic talk shows, recitations, livestreaming and debates. At the post-training stage, we curate a diverse instruction-tuning corpus and introduce thinking mechanisms into both audio understanding and generation. MiMo-Audio-7B-Instruct achieves open-source SOTA on audio understanding benchmarks, spoken dialogue benchmarks and instruct-TTS evaluations, approaching or surpassing closed-source models.

<p align="center"> <img width="95%" src="https://github.com/XiaomiMiMo/MiMo-Audio/blob/main/assets/Results.png?raw=true"> </p>

Architecture

MiMo-Audio-Tokenizer

MiMo-Audio-Tokenizer is a 1.2B-parameter Transformer operating at 25 Hz. It employs an eight-layer RVQ stack to generate 200 tokens per second. By jointly optimizing semantic and reconstruction objectives, we train MiMo-Audio-Tokenizer from scratch on a 10-million-hour corpus, achieving superior reconstruction quality and facilitating downstream language modeling.

<p align="center"> <img width="95%" src="https://github.com/XiaomiMiMo/MiMo-Audio/blob/main/assets/tokenizer.png?raw=true"> </p>

MiMo-Audio couples a patch encoder, an LLM, and a patch decoder to improve modeling efficiency for high-rate sequences and bridge the length mismatch between speech and text. The patch encoder aggregates four consecutive time steps of RVQ tokens into a single patch, downsampling the sequence to a 6.25 Hz representation for the LLM. The patch decoder autoregressively generates the full 25 Hz RVQ token sequence via a delayed-generation scheme.

MiMo-Audio

<p align="center"> <img width="95%" src="https://github.com/XiaomiMiMo/MiMo-Audio/blob/main/assets/architecture.png?raw=true"> </p>

Explore MiMo-Audio Now! 🚀🚀🚀

• 📰 Read the Official Blog: MiMo-Audio Blog
• 📄 Dive into the Technical Report: MiMo-Audio Technical Report

Model Download

pip install huggingface-hub

hf download XiaomiMiMo/MiMo-Audio-Tokenizer --local-dir ./models/MiMo-Audio-Tokenizer
hf download XiaomiMiMo/MiMo-Audio-7B-Base --local-dir ./models/MiMo-Audio-7B-Base
hf download XiaomiMiMo/MiMo-Audio-7B-Instruct --local-dir ./models/MiMo-Audio-7B-Instruct

Getting Started

Spin up the MiMo-Audio demo in minutes with the built-in Gradio app.

Prerequisites (Linux)

• Python 3.12
• CUDA >= 12.0

Installation

git clone https://github.com/XiaomiMiMo/MiMo-Audio.git
cd MiMo-Audio
pip install -r requirements.txt
pip install flash-attn==2.7.4.post1

[!Note] If the compilation of flash-attn takes too long, you can download the precompiled wheel and install it manually:

• Download Precompiled Wheel

pip install /path/to/flash_attn-2.7.4.post1+cu12torch2.6cxx11abiFALSE-cp312-cp312-linux_x86_64.whl

Run the demo

python run_mimo_audio.py

This launches a local Gradio interface where you can try MiMo-Audio interactively.

<p align="center"> <img width="95%" src="https://github.com/XiaomiMiMo/MiMo-Audio/blob/main/assets/demo_ui.jpg?raw=true"> </p>

Enter the local paths for MiMo-Audio-Tokenizer and MiMo-Audio-7B-Instruct, then enjoy the full functionality of MiMo-Audio!

Inference Scripts

Base Model

We provide an example script to explore the in-context learning capabilities of MiMo-Audio-7B-Base.
See: inference_example_pretrain.py

Instruct Model

To try the instruction-tuned model MiMo-Audio-7B-Instruct, use the corresponding inference script.
See: inference_example_sft.py

Evaluation Toolkit

Full evaluation suite are available at 🌐MiMo-Audio-Eval.

This toolkit is designed to evaluate MiMo-Audio and other recent audio LLMs as mentioned in the paper. It provides a flexible and extensible framework, supporting a wide range of datasets, tasks, and models.

Citation

@misc{coreteam2025mimoaudio,
      title={MiMo-Audio: Audio Language Models are Few-Shot Learners}, 
      author={LLM-Core-Team Xiaomi},
      year={2025},
      url={GitHub - XiaomiMiMo/MiMo-Audio}, 
}

Contact

Please contact us at [email protected] or open an issue if you have any questions.