Skip to content
ModelListedExternal
Download

LibreOVDEIMs

OV-DEIM-S real-time open-vocabulary detector (11M params, ViT-tiny (DEIMv2 distilled) backbone) repackaged for LibreYOLO's LibreOpenVocab tier, bundled with the MobileCLIP-B(LT) text tower for online prompt encoding.

Sourcehuggingface.co/LibreYOLO/LibreOVDEIMs/tree/6d34cfa5ab63f14f540ec1071dd8e9201240e5d8
KindText modelVersionmodelBgrade

Model source

Kind
Text model
File size
285 MB
Version
vmodel
Updated
Oct 21, 2026
Downloads
33
Source
Hugging Face

Source description

LibreOVDEIMs

OV-DEIM-S real-time open-vocabulary detector (11M params, ViT-tiny (DEIMv2 distilled) backbone) repackaged for LibreYOLO's LibreOpenVocab tier, bundled with the MobileCLIP-B(LT) text tower for online prompt encoding.

Source

Detector converted from wleilei/OV-DEIM at commit dfbf394672407b7f837ec08e7d68e8127548b254 (checkpoint ovdeim_s.pth, trained on Objects365v1 + GoldG). Text encoder taken from apple/MobileCLIP-B-LT-OpenCLIP.

Paper: OV-DEIM: Real-time DETR-Style Open-Vocabulary Object Detection with GridSynthetic Augmentation (arXiv 2603.07022). Upstream reports 161 FPS on a T4 with TensorRT at 640 input.

Modifications

State-dict key remapping and merge only (module. prefix stripped, the training-only denoising embedding dropped, the MobileCLIP text tower stored under text_encoder.*). Learned parameters are unchanged. See weights/convert_ovdeim_weights.py in the LibreYOLO source repository.

LibreYOLO's port reproduces the upstream paper's zero-shot COCO val2017 result through this artifact (45.9 AP for the L size).

Usage

from libreyolo import LibreOpenVocab

model = LibreOpenVocab("ov-deim-s")
model.set_classes(["person", "a red backpack", "forklift"])
results = model.predict("image.jpg")

License

Layered, most restrictive governs the combined artifact:

  • Detector weights: CC BY-NC 4.0 (non-commercial, attribution; redistribution and format conversion permitted by upstream's MODEL_LICENSE).
  • OV-DEIM code lineage: Apache-2.0.
  • MobileCLIP-B(LT) text tower: Apple Machine Learning Research Model license, research use only.
  • DINOv3-derived backbone: DINOv3 License.

See NOTICE for attribution details.

Sources

1 source
Hugging FacePrimaryVersion 6d34cfa5ab63f14f540ec1071dd8e9201240e5d8
huggingface.co/LibreYOLO/LibreOVDEIMs/tree/6d34cfa5ab63f14f540ec1071dd8e9201240e5d8

Verified Oct 21

Model artifacts

1 artifact

model.safetensors

safetensors · 285 MB · SHA-256 5948e74c37db…b0af · Hugging Face

Download file

Source excerpts

3 excerpts
ReadmeAug 5huggingface.co/LibreYOLO/LibreOVDEIMs

--- license: cc-by-nc-4.0 library_name: libreyolo tags: - object-detection - zero-shot-object-detection - open-vocabulary-detection - libreyolo --- # LibreOVDEIMs OV-DEIM-S real-time open-vocabulary detector (11M params, ViT-tiny (DEIMv2 distilled) backbone) repackaged for LibreYOLO's `LibreOpenVocab` tier, bundled with the MobileCLIP-B(LT) text tower for online prompt encoding. ## Source Detector converted from [wleilei/OV-DEIM](https://github.com/wleilei/OV-DEIM) at commit `dfbf394672407b7f837ec08e7d68e8127548b254` (checkpoint `ovdeim_s.pth`, trained on Objects365v1 + GoldG). Text encoder taken from [apple/MobileCLIP-B-LT-OpenCLIP](https://huggingface.co/apple/MobileCLIP-B-LT-OpenCLIP). Paper: OV-DEIM: Real-time DETR-Style Open-Vocabulary Object Detection with GridSynthetic Augmentation (arXiv 2603.07022). Upstream reports 161 FPS on a T4 with TensorRT at 640 input. ## Modifications State-dict key remapping and merge only (`module.` prefix stripped, the training-only denoising embedding dropped, the MobileCLIP text tower stored under `text_encoder.*`). Learned parameters are unchanged. See `weights/convert_ovdeim_weights.py` in the [LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo). LibreYOLO's port reproduces the upstream paper's zero-shot COCO val2017 result through this artifact (45.9 AP for the L size). ## Usage ```python from libreyolo import LibreOpenVocab model = LibreOpenVocab("ov-deim-s") model.set_classes(["person", "a red backpack", "forklift"]) results = model.predict("image.jpg") ``` ## License Layered, most restrictive governs the combined artifact: - Detector weights: [CC BY-NC 4.0](./LICENSE) (non-commercial, attribution; redistribution and format conversion permitted by upstream's MODEL_LICENSE). - OV-DEIM code lineage: [Apache-2.0](./LICENSE_CODE). - MobileCLIP-B(LT) text tower: [Apple Machine Learning Research Model license](./LICENSE_MOBILECLIP_WEIGHTS), research use only. - DINOv3-derived backbone: [DINOv3 License](./LICENSE_DINOV3). See [`NOTICE`](./NOTICE) for attribution details.

NoteOct 21huggingface.co/LibreYOLO/LibreOVDEIMs

Source context: 33 downloads · 0 likes · Pipeline object-detection · Library libreyolo · Repo LibreYOLO/LibreOVDEIMs

NoteAug 5huggingface.co/LibreYOLO/LibreOVDEIMs

Source context: 28 downloads · 0 likes · Pipeline object-detection · Library libreyolo · Repo LibreYOLO/LibreOVDEIMs