Instructions to use AesSedai/MiMo-V2.5-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AesSedai/MiMo-V2.5-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="AesSedai/MiMo-V2.5-GGUF",
	filename="IQ3_S/MiMo-V2.5-IQ3_S-00001-of-00004.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use AesSedai/MiMo-V2.5-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Use Docker

docker model run hf.co/AesSedai/MiMo-V2.5-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use AesSedai/MiMo-V2.5-GGUF with Ollama:
```
ollama run hf.co/AesSedai/MiMo-V2.5-GGUF:Q4_K_M
```

Unsloth Studio

How to use AesSedai/MiMo-V2.5-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AesSedai/MiMo-V2.5-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for AesSedai/MiMo-V2.5-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for AesSedai/MiMo-V2.5-GGUF to start chatting

How to use AesSedai/MiMo-V2.5-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "AesSedai/MiMo-V2.5-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use AesSedai/MiMo-V2.5-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use AesSedai/MiMo-V2.5-GGUF with Docker Model Runner:
```
docker model run hf.co/AesSedai/MiMo-V2.5-GGUF:Q4_K_M
```

Lemonade

How to use AesSedai/MiMo-V2.5-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AesSedai/MiMo-V2.5-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.MiMo-V2.5-GGUF-Q4_K_M

List all available models

lemonade list

MiMo-V2.5-GGUF

File size: 2,726 Bytes

eed9c5e

---
base_model:
- XiaomiMiMo/MiMo-V2.5
---
## Notes
- 05/08/26: The CUDA flash attention branch has also been merged to master, please use the master branch and recompile!
- 05/07/26: The PR branch has been merged to master. All of the quants and imatrix have been updated with the newest conversion and are ready to re-download. These quants include MTP tensors for when that gets added upstream eventually.
- 05/05/26: ~~I've updated all the quants to use the fused QKV conversion. The PR branch supports both fused + unfused so it's not necessary to download the new quants, but it may provide a small speed boost.~~
- 05/03/26: WIP vision support on this branch: https://github.com/AesSedai/llama.cpp/tree/mimo-v2.5-vision (if it's broken with F16 mmproj, pull the latest commit and recompile, or try the BF16 mmproj) and uploaded mmproj files
- 05/01/26: ~~This branch includes CUDA flash attention, should speed up PP / TG: https://github.com/AesSedai/llama.cpp/tree/mimo-v2.5-fattn~~
- 04/28/26: ~~I recommend pulling and compiling from this PR branch to run the model: https://github.com/ggml-org/llama.cpp/pull/22493.~~

## Model
This is a text-only GGUF quantization of XiaomiMiMo/MiMo-V2.5. This means that image and audio input is not present in this GGUF, and will not be available until support is added upstream in llama.cpp.

This repo contains specialized MoE-quants for MiMo-V2.5. The idea being that given the huge size of the FFN tensors compared to the rest of the tensors in the model, it should be possible to achieve a better quality while keeping the overall size of the entire model smaller compared to a similar naive quantization. To that end, the quantization type default is kept in high quality and the FFN UP + FFN GATE tensors are quanted down along with the FFN DOWN tensors.

| Quant | Size | Mixture | PPL | 1-(Mean PPL(Q)/PPL(base)) | KLD |
| :--------- | :--------- | :------- | :------- | :------- | :------- |
| Q8_0 | 306.66 GiB (8.50 BPW) | Unknown / TBD | 5.134769 ± 0.030261 | +0.1230% | 0.012010 ± 0.000150 |
| Q5_K_M | 213.39 GiB (5.92 BPW) | Q8_0 / Q5_K / Q5_K / Q6_K | 5.147654 ± 0.030377 | +0.3743% | 0.014752 ± 0.000240 |
| Q4_K_M | 177.68 GiB (4.93 BPW) | Q8_0 / Q4_K / Q4_K / Q5_K | 5.202785 ± 0.030828 | +1.4493% | 0.020631 ± 0.000251 |
| IQ4_XS | 137.75 GiB (3.82 BPW) | Q8_0 / IQ3_S / IQ3_S / IQ4_XS | 5.272594 ± 0.031193 | +2.8105% | 0.041508 ± 0.000343 |
| IQ3_S | 106.31 GiB (2.95 BPW) | Q6_K / IQ2_S / IQ2_S / IQ3_S | 5.545001 ± 0.033188 | +8.1221% | 0.092415 ± 0.000600 |

![kld_graph](kld_data/01_kld_vs_filesize.png "Chart showing Pareto KLD analysis of quants")
![ppl_graph](kld_data/02_ppl_vs_filesize.png "Chart showing Pareto PPL analysis of quants")