Multimodal AI is rapidly evolving, MiniCPM-V 4.0 by OpenBMB emerges as a game-changer, combining cutting-edge visual understanding with unprecedented efficiency. Built on SigLIP2-400M and MiniCPM4-3B, this compact yet powerful model packs 4.1B parameters, but consistently punches above its weight. It not only inherits the strong single-image, multi-image, and video comprehension capabilities of its predecessor (MiniCPM-V 2.6), but also surpasses it with remarkable efficiency. Benchmark results on OpenCompass demonstrate this leap. MiniCPM-V 4.0 achieves a 69.0 average score, outperforming models like GPT-4.1-mini-20250414, MiniCPM-V 2.6 (8.1B), and Qwen2.5-VL-3B-Instruct, proving that smaller can indeed be smarter. What makes it even more exciting is its real-world usability: the model runs seamlessly on end devices, delivering under 2s first-token delay and over 17 tokens/s decoding on iPhone 16 Pro Max, all without heating issues, making on-device multimodal AI finally practical. With easy integration across frameworks like llama.cpp, Ollama, vLLM, SGLang, LLaMA-Factory, and even a native iOS app, MiniCPM-V 4.0 isn’t just another AI model, it’s a versatile, efficient, and deployment-ready multimodal powerhouse.
In this article, we’re going to see a step-by-step process to install and run this model locally or in GPU-accelerated environments.
Prerequisites
The minimum system requirements for running this model are:
- GPU: 1x RTX4090 or 1x RTX A6000
- Storage: 50 GB (preferable)
- VRAM: at least 16 GB
- Anaconda installed
Step-by-step process to install and run MiniCPM-v4
For the purpose of this tutorial, we’ll use a GPU-powered Virtual Machine by NodeShift since it provides high compute Virtual Machines at a very affordable cost on a scale that meets GDPR, SOC2, and ISO27001 requirements. Also, it offers an intuitive and user-friendly interface, making it easier for beginners to get started with Cloud deployments. However, feel free to use any cloud provider of your choice and follow the same steps for the rest of the tutorial.
Step 1: Setting up a NodeShift Account
Visit app.nodeshift.com and create an account by filling in basic details, or continue signing up with your Google/GitHub account.
If you already have an account, login straight to your dashboard.
Step 2: Create a GPU Node
After accessing your account, you should see a dashboard (see image), now:
- Navigate to the menu on the left side.
- Click on the GPU Nodes option.
- Click on Start to start creating your very first GPU node.
These GPU nodes are GPU-powered virtual machines by NodeShift. These nodes are highly customizable and let you control different environmental configurations for GPUs ranging from H100s to A100s, CPUs, RAM, and storage, according to your needs.
Step 3: Selecting configuration for GPU (model, region, storage)
- For this tutorial, we’ll be using 1x RTX A6000 GPU, however, you can choose any GPU as per the prerequisites.
- Similarly, we’ll opt for 200GB storage by sliding the bar. You can also select the region where you want your GPU to reside from the available ones.
Step 4: Choose GPU Configuration and Authentication method
- After selecting your required configuration options, you’ll see the available GPU nodes in your region and according to (or very close to) your configuration. In our case, we’ll choose a 1x RTX A6000 48GB GPU node with 64vCPUs/63GB RAM/200GB SSD.
2. Next, you’ll need to select an authentication method. Two methods are available: Password and SSH Key. We recommend using SSH keys, as they are a more secure option. To create one, head over to our official documentation.
Step 5: Choose an Image
The final step is to choose an image for the VM, which in our case is Nvidia Cuda.
That’s it! You are now ready to deploy the node. Finalize the configuration summary, and if it looks good, click Create to deploy the node.
Step 6: Connect to active Compute Node using SSH
- As soon as you create the node, it will be deployed in a few seconds or a minute. Once deployed, you will see a status Running in green, meaning that our Compute node is ready to use!
- Once your GPU shows this status, navigate to the three dots on the right, click on Connect with SSH, and copy the SSH details that appear.
As you copy the details, follow the below steps to connect to the running GPU VM via SSH:
- Open your terminal, paste the SSH command, and run it.
2. In some cases, your terminal may take your consent before connecting. Enter ‘yes’.
3. A prompt will request a password. Type the SSH password, and you should be connected.
Output:
Next, If you want to check the GPU details, run the following command in the terminal:
!nvidia-smi
Step 7: Set up the project environment with dependencies
- Create a virtual environment using Anaconda.
conda create -n minicpm python=3.11 -y && conda activate minicpm
Output:
2. Install required dependencies.
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
pip install einops timm pillow
pip install git+https://github.com/huggingface/transformers
pip install git+https://github.com/huggingface/accelerate
pip install git+https://github.com/huggingface/diffusers
pip install huggingface_hub
pip install sentencepiece bitsandbytes protobuf decord numpy
Output:
3. Install and run jupyter notebook.
conda install -c conda-forge --override-channels notebook -y
conda install -c conda-forge --override-channels ipywidgets -y
jupyter notebook --allow-root
4. If you’re on a remote machine (e.g., NodeShift GPU), you’ll need to do SSH port forwarding in order to access the jupyter notebook session on your local browser.
Run the following command in your local terminal after replacing:
<YOUR_SERVER_PORT> with the PORT allotted to your remote server (For the NodeShift server – you can find it in the deployed GPU details on the dashboard).
<PATH_TO_SSH_KEY> with the path to the location where your SSH key is stored.
<YOUR_SERVER_IP> with the IP address of your remote server.
ssh -L 8888:localhost:8888 -p <YOUR_SERVER_PORT> -i <PATH_TO_SSH_KEY> root@<YOUR_SERVER_IP>
Output:
After this copy the URL you received in your remote server:
And paste this on your local browser to access the Jupyter Notebook session.
Step 8: Download and Run the model
- Open a Python notebook inside Jupyter.
2. Download the model checkpoints.
from PIL import Image
import torch
from transformers import AutoModel, AutoTokenizer
model_path = 'openbmb/MiniCPM-V-4'
model = AutoModel.from_pretrained(model_path, trust_remote_code=True,
# sdpa or flash_attention_2, no eager
attn_implementation='sdpa', torch_dtype=torch.bfloat16)
model = model.eval().cuda()
tokenizer = AutoTokenizer.from_pretrained(
model_path, trust_remote_code=True)
image = Image.open('./landform.jpg').convert('RGB')
# First round chat
question = "What is the landform in the picture?"
msgs = [{'role': 'user', 'content': [image, question]}]
answer = model.chat(
msgs=msgs,
image=image,
tokenizer=tokenizer
)
print(answer)
# Second round chat, pass history context of multi-turn conversation
msgs.append({"role": "assistant", "content": [answer]})
msgs.append({"role": "user", "content": [
"What should I pay attention to when traveling here?"]})
answer = model.chat(
msgs=msgs,
image=None,
tokenizer=tokenizer
)
print(answer)
Output:
Here’s the image we used to testing the model:
Picsum ID: 866
Output:
Conclusion
To wrap up, MiniCPM-V 4.0 clearly demonstrates how multimodal AI is becoming more efficient, accessible, and deployment-ready, setting a new benchmark in balancing compact design with powerful visual and reasoning capabilities. From its ability to outperform larger models on benchmarks to its seamless real-world usability on devices like the iPhone 16 Pro Max, it proves that high performance no longer requires massive scale. At the same time, Nodeshift Cloud makes experimenting with and deploying such state-of-the-art models far more practical, offering GPU-accelerated environments, simple setup workflows, and flexible scaling to match your needs.
