mimo-v2.6-distill-qwen-9b-torch

Agentic 9B vision-language model distilled from MiMo-V2.6 onto Qwen3.5 for object detection via 2D grounding.

Details

• Model name: mimo-v2.6-distill-qwen-9b-torch

• Model source: https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B

• Model author: Xiaomi MiMo

• Model license: MIT

• Model size: 17.56 GB

• Exposes embeddings? no

• Tags: detection, vlm, torch, transformer, zero-shot

Requirements

• Packages: torch, torchvision, transformers>=5.12.0, accelerate

• CPU support

  • yes

• GPU support

  • yes

Example usage

import fiftyone as fo
import fiftyone.zoo as foz

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("mimo-v2.6-distill-qwen-9b-torch")

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

session = fo.launch_app(dataset)