Qwen/Qwen2.5-VL-32B-Instruct

Note

This is a remotely-sourced model from the qwen2_5_vl plugin, maintained by the community. It is not part of FiftyOne core and may have special installation requirements. Please review the plugin documentation and license before use.

Qwen2.5-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud..

Details

• Model name: Qwen/Qwen2.5-VL-32B-Instruct

• Model source: https://huggingface.co/Qwen/Qwen2.5-VL-32B-Instruct

• Model author: QwenLM

• Model license: Apache-2.0 license

• Exposes embeddings? no

• Tags: detection, segmentation, ocr, VLM, zero-shot

Requirements

• Packages: huggingface-hub, transformers, torch, torchvision, qwen-vl-utils, accelerate

• CPU support

  • yes

• GPU support

  • yes

Example usage

import fiftyone as fo
import fiftyone.zoo as foz

foz.register_zoo_model_source("https://github.com/harpreetsahota204/qwen2_5_vl")

dataset = foz.load_zoo_dataset(
    "coco-2017",
    split="validation",
    dataset_name=fo.get_default_dataset_name(),
    max_samples=50,
    shuffle=True,
)

model = foz.load_zoo_model("Qwen/Qwen2.5-VL-32B-Instruct")

dataset.apply_model(model, label_field="predictions")

session = fo.launch_app(dataset)