# NVIDIA LocateAnything-3B — Parallel Box Decoding Vision-Language Grounder Hits 12.7 Boxes/Sec on H100, ~10× Faster Than Qwen3-VL, Trained on 138M Queries Across 785M Boxes

> LocateAnything decodes full bounding boxes in one shot, getting NVIDIA a vision-language grounder that is roughly 10× faster than Qwen3-VL.

- Org: NVIDIA
- Date: 2026-05-26
- Category: model, paper, repo
- Source: https://huggingface.co/nvidia/LocateAnything-3B
- Canonical page: https://ai-tldr.dev/releases/nvidia-locateanything-3b/

## In short

A 3B vision-language grounding model that emits whole bounding boxes in parallel instead of token-by-token. Apache 2.0 code, NVIDIA-licensed weights, 138M-query training set, and a live HF Space demo.

## What it is

A 3B vision-language model from NVIDIA Labs for visual grounding: open-set detection, dense detection, referring-expression grounding, GUI element grounding, OCR localization, and document layout. Built on a MoonViT vision encoder plus Qwen2.5-3B-Instruct language model.

## How it works

The novelty is Parallel Box Decoding (PBD), which treats a bounding box as an atomic unit and predicts its four coordinates in one forward pass instead of serializing coordinates into a token stream. Three inference modes (fast / slow / hybrid) trade off speed against robustness. Trained on LocateAnything-Data: 12M images, 138M queries, 785M boxes, labelled by humans plus Qwen3-VL, Molmo, and SAM 3.

## Why it matters

Autoregressive VLMs cap throughput because every coordinate is another decoding step. PBD turns localization into a parallel-friendly task, unlocking real-time grounding for robotics, GUI agents, and document AI without giving up accuracy. Hits LVIS F1@Mean 50.7 and ScreenSpot-Pro 60.3 mean F1 — competitive with much larger models.

## Who it's for

Researchers and product teams building visual grounding into GUI agents, robotics perception, autonomous driving, or document understanding.

## Try it

pip install transformers, then load `nvidia/LocateAnything-3B` and call `generate(..., generation_mode="hybrid")`.

## Links

- [Model Card](https://huggingface.co/nvidia/LocateAnything-3B)
- [Project Page](https://research.nvidia.com/labs/lpr/locate-anything/)
- [Paper (arXiv)](https://arxiv.org/abs/2605.27365)
- [Code (NVlabs/Eagle)](https://github.com/NVlabs/Eagle)
- [Live Demo](https://huggingface.co/spaces/nvidia/LocateAnything)