QVQ 72B Preview Abliterated GPTQ Int8

🚀 QVQ-72B-Preview-abliterated-GPTQ-Int8

This is a GPTQ-quantized 8-bit version of huihui-ai/QVQ-72B-Preview-abliterated, offering a toolkit for convenient handling of various visual inputs.

🚀 Quick Start

This project provides a GPTQ-quantized 8-bit version of the model, along with a toolkit to handle different types of visual inputs.

✨ Features

• GPTQ quantization to 8-bit, which can effectively reduce model size and inference time.
• Support for various visual input types, including base64, URLs, and interleaved images and videos.
• Provide a convenient toolkit to simplify the processing of visual inputs.

📦 Installation

You can install the necessary toolkit using the following command:

pip install qwen-vl-utils

💻 Usage Examples

Basic Usage

The following is a code example demonstrating how to use the chat model with transformers and qwen_vl_utils:

from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info

# default: Load the model on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained(
    "huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8", torch_dtype="auto", device_map="auto"
)

# default processer
processor = AutoProcessor.from_pretrained("huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8")

# The default range for the number of visual tokens per image in the model is 4-16384. You can set min_pixels and max_pixels according to your needs, such as a token count range of 256-1280, to balance speed and memory usage.
# min_pixels = 256*28*28
# max_pixels = 1280*28*28
# processor = AutoProcessor.from_pretrained("huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8", min_pixels=min_pixels, max_pixels=max_pixels)

messages = [
    {
        "role": "system",
        "content": [
            {"type": "text", "text": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."}
        ],
    },
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/QVQ/demo.png",
            },
            {"type": "text", "text": "What value should be filled in the blank space?"},
        ],
    }
]

# Preparation for inference
text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
    text=[text],
    images=image_inputs,
    videos=video_inputs,
    padding=True,
    return_tensors="pt",
)
inputs = inputs.to("cuda")

# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=8192)
generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text)

📄 License

This project uses the Qwen License.