Minicpm O 2 6 Int4

🚀 MiniCPM-o 2.6 int4: A GPT-4o Level MLLM for Vision, Speech and Multimodal Live Streaming on Your Phone

MiniCPM-o 2.6 int4 is an int4 quantized version of MiniCPM-o 2.6, which can significantly reduce GPU memory usage while maintaining high performance for vision, speech, and multimodal live - streaming tasks.

📦 Installation

Prepare code and install AutoGPTQ

We are submitting a PR to officially support minicpm-o 2.6 inference. You can follow these steps to set up the environment:

git clone https://github.com/OpenBMB/AutoGPTQ.git && cd AutoGPTQ
git checkout minicpmo

# install AutoGPTQ
pip install -vvv --no-build-isolation -e .

💻 Usage Examples

Basic Usage

This is how you can use the MiniCPM-o-2_6-int4 model. You need to change the model initialization part to AutoGPTQForCausalLM.from_quantized.

import torch
from transformers import AutoModel, AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM

model = AutoGPTQForCausalLM.from_quantized(
    'openbmb/MiniCPM-o-2_6-int4',
    torch_dtype=torch.bfloat16,
    device="cuda:0",
    trust_remote_code=True,
    disable_exllama=True,
    disable_exllamav2=True
)
tokenizer = AutoTokenizer.from_pretrained(
    'openbmb/MiniCPM-o-2_6-int4',
    trust_remote_code=True
)

model.init_tts()

For more detailed usage reference, please check MiniCPM-o-2_6#usage.

📄 Model Information