Allen8/TVC-7B

Model Summary

The TVC models are 7B parameter models based on Qwen2-VL-7B-Instruct model with a context window of 8K tokens.

• Repository: https://github.com/sun-hailong/TVC
• Project Page: https://sun-hailong.github.io/projects/TVC/
• Languages: English, Chinese
• Paper: https://arxiv.org/abs/2503.13360

Model Architecture

• Architecture: Qwen2-VL-7B-Instruct
• Data: a mixture of 300k long-chain reasoning data
• Precision: BFloat16

Hardware & Software

• Hardware: 64 * NVIDIA Tesla H20
• Orchestration: HuggingFace Trainer
• Code: Pytorch

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3

Quick Start

from vllm import LLM, SamplingParams
from PIL import Image

model_name = "Allen8/TVC-72B"
llm = LLM(
        model=model_name,
        trust_remote_code=True,
        tensor_parallel_size=8,
    )

question = "Hint: Please answer the question requiring an integer answer and provide the final value, e.g., 1, 2, 3, at the end.
Question: Subtract all red things. Subtract all tiny matte balls. How many objects are left?
Please answer the question using a long-chain reasoning style and think step by step."
placeholder = "<|image_pad|>"
prompt = ("<|im_start|>system
You are a helpful assistant.<|im_end|>
"
f"<|im_start|>user
<|vision_start|>{placeholder}<|vision_end|>"
f"{question}<|im_end|>
"
"<|im_start|>assistant
")

sampling_params = SamplingParams(
    temperature=0.0,
    top_k=1,
    top_p=1.0,
    stop_token_ids=[],
    repetition_penalty=1.05,
    max_tokens=8192
)

image = Image.open("images/case1.png")
inputs = {
            "prompt": prompt,
            "multi_modal_data": {
                "image": image
            },
        }

outputs = llm.generate([inputs], sampling_params=sampling_params)
print(outputs[0].outputs[0].text)

Citation

@article{sun2024mitigating,
    title={Mitigating Visual Forgetting via Take-along Visual Conditioning for Multi-modal Long CoT Reasoning},
    author={Sun, Hai-Long and Sun, Zhun and Peng, Houwen and Ye, Han-Jia},
    journal={arXiv preprint arXiv:2503.13360},
    year={2025}
}