---
title: IDEA-Research/grounding-dino-tiny
description: "The Grounding DINO model was proposed in Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection by Shilong L. Swin, 172M parameters."
canonical_url: https://superlinked.com/models/idea-research-grounding-dino-tiny
last_updated: 2026-09-05
---

# IDEA-Research/grounding-dino-tiny

The Grounding DINO model was proposed in Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.

Source: [IDEA-Research/grounding-dino-tiny on HuggingFace](https://huggingface.co/IDEA-Research/grounding-dino-tiny)

## Overview

| Field | Value |
|-------|-------|
| Architecture | Swin |
| Parameters | 172M |
| Tasks | Extract |
| Outputs | Bounding Boxes |
| License | apache-2.0 |
| Inputs | text, image |

## Benchmarks

### COCO

Domain: general · Task: detection · Language: en

Object detection on COCO natural images

Corpus: 5,000 · Queries: 5,000

**Variant: default_limit-1000**

**Quality:** ap: 0.4315 · ap50: 0.5859 · ap75: 0.4638 · ar 100: 0.5291

**Performance (A10G b1 c4):** Detect 0.0 mpix/s · Detect p50 43.7s

**Performance (L4-SPOT b1 c4):** Detect 0.9 mpix/s · Detect p50 734.4ms

**Performance (L4 b1 c4):** Detect 0.9 mpix/s · Detect p50 330.7ms

**Performance (L4 b1 c16):** Detect 1.1 mpix/s · Detect p50 263.8ms

**Variant: default_limit-100**

**Quality:** ap: 0.4860 · ap50: 0.6553 · ap75: 0.5001 · ar 100: 0.5593

**Performance (RTX-4090 b1 c16):** Detect 4.0 mpix/s · Detect p50 602.2ms

**Variant: default**

**Quality:** ap: 0.4028 · ap50: 0.5627 · ap75: 0.4361 · ar 100: 0.5211

**Performance (RTX-PRO-6000-Blackwell-Server-Edition b1 c16):** Detect 2.6 mpix/s · Detect p50 113.9ms

[Reference](https://cocodataset.org/)