LocateAnything

🔗 Quick Links

• 🚀 Online Demo: LocateAnything (Hugging Face Spaces)
• 💻 GitHub Code: NVlabs/Eagle/Embodied
• 📄 Paper: arXiv:2605.27365

Model Overview

Description:

LocateAnything is a vision-language model for fast and high-quality visual grounding, enabling precise object localization, dense detection, and point-based localization across diverse domains in both Enterprise Intelligence and Physical AI. The model adopts a generalist design, supporting tasks such as referring expression grounding, multi-object detection, GUI element grounding, and text localization, with strong performance in complex and cluttered scenes.

Its core innovation, Parallel Box Decoding (PBD), predicts complete bounding box coordinates in a single parallel step rather than autoregressive token-by-token decoding, improving efficiency while preserving geometric consistency. This enables up to 2.5× higher throughput compared to prior approaches.

The model is trained on a large-scale multi-domain dataset (12M images, 138M+ queries, 785M bounding boxes) spanning natural scenes, robotics, driving, GUI interaction, and document understanding. It serves as a foundation for generalist multimodal perception and has been integrated into NVIDIA’s frontier production-grade vision-language models, such as Nemotron 3 Nano Omni, supporting grounding, GUI understanding, and multimodal agentic capabilities.

LocateAnything is developed as part of the Eagle VLM model family. This released model is for research and development only. In addition, LocateAnything contributed to Nemotron and Cosmos as part of the Computer Use and Visual Grounding features. We give special thanks the Nemotron and Cosmos Teams for time and efforts in product integration.

Demo Videos

Your browser does not support the video tag.

Your browser does not support the video tag.

License/Terms of Use:

This model is released under the NVIDIA License for non-commercial use, which permits use, reproduction, and modification for academic and non-profit research purposes only. Commercial use is not permitted, except by NVIDIA and its affiliates. Redistribution must retain the license and all applicable copyright and attribution notices. The model is provided “as is” without warranty of any kind, and users assume all associated risks.

This model is built using components from third-party models with their respective licenses:

• Language model: Qwen2.5-3B-Instruct (Qwen Research License)
• Vision encoder: MoonViT-SO-400M (MIT License)

Models are improved using Qwen.

Deployment Geography:

Global

Use Case:

LocateAnything-3B is intended for developers and researchers building vision-language models and applications that require fast and precise visual localization from natural language instructions.

Supported use cases include:

• Open-set, common, and long-tail object detection
• Dense multi-object detection in cluttered scenes
• Phrase and referring-expression grounding
• Automated dataset labeling and annotation (e.g., detection, grounding, pointing)
• GUI element grounding for interactive and agentic systems
• Robotics and autonomous driving perception
• Document understanding, layout grounding, and OCR localization
• Industrial inspection, surveillance, and remote sensing applications
• Point-based localization and fine-grained spatial reasoning

Release Date [Insert the expected release date below]:

• Github [05/26/2026] via https://github.com/NVlabs/Eagle/tree/main/Embodied.
• Hugging Face [05/26/2026] via https://huggingface.co/nvidia/LocateAnything-3B.
• Demo [05/26/2026] via https://huggingface.co/spaces/nvidia/LocateAnything.
• Webpage [05/26/2026] via https://research.nvidia.com/labs/lpr/locate-anything/.
• Tech Report [05/26/2026] via https://research.nvidia.com/labs/lpr/locate-anything/LocateAnything.pdf

References(s):

• Wang et al., LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding, NVIDIA Tech Report, 2026
• Kimi Team, Kimi-VL Technical Report, arXiv:2504.07491, 2025.
• Qwen Team, Qwen2.5: A Party of Foundation Models, Qwen Blog, 2024.
• Chen et al., Pix2Seq: A Language Modeling Framework for Object Detection, ICLR, 2022.
• Jiang et al., Detect Anything via Next Point Prediction, arXiv:2510.12798, 2025.
• Liu et al., Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection, arXiv:2303.05499, 2023.
• Lin et al., Microsoft COCO: Common Objects in Context, ECCV, 2014.
• Gupta et al., LVIS: A Dataset for Large Vocabulary Instance Segmentation, CVPR, 2019.
• Li et al., ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use, ACM MM, 2025.

Model Architecture:

Architecture Type: Transformer-based vision-language model (VLM).

Network Architecture: Native-resolution VLM with the following components:

• Vision encoder: MoonViT
• Language model: Qwen2.5-3B-Instruct
• Multimodal projector: MLP projector
• Output formulation: Block-based structure for visual grounding

Number of model parameters: 3B.

LocateAnything extends a vision-language model with Parallel Box