> This is the documentation for LibreYOLO 1.6.0. Current version: https://www.libreyolo.com/docs/models/northmicrovision

# North Micro Vision
North Micro Vision detects objects through a text vocabulary.
Tasks: Detection. Install: pip install "libreyolo[vlm]".
Verified against LibreYOLO v1.6.0.

## Install

```bash
pip install "libreyolo[vlm]" "transformers>=5.16.0"
```

## Predict

**Python**

```python
from libreyolo import LibreVLM, SAMPLE_IMAGE

model = LibreVLM("north-micro-vision", device="cpu")
model.set_classes(["person", "building"])
result = model(SAMPLE_IMAGE)
print(result.boxes)
```

The adapter runs one detection query per vocabulary class. It requires Transformers 5.16 or later, above the shared VLM extra floor. Training is not supported.

## Licensing

Check the license on the Hugging Face repository of the specific weights you download. That repository is authoritative and licenses are not always uniform across a family. This is a description of the licenses involved, not legal advice.

- Original work: North Micro Vision, Cohere
- Upstream license: Apache-2.0
- Upstream source: https://huggingface.co/CohereLabs
- LibreYOLO code: MIT
- Weights: Apache-2.0, distributed by their authors. LibreYOLO does not host or mirror them.
- Interpretation: The model publisher declares Apache-2.0.


## North Micro Vision architecture

- North Micro Vision 2.4B: [SVG](/diagrams/models/northmicrovision/2-4b-detect.svg), [interactive diagram](/diagrams/models/northmicrovision/2-4b-detect.html)