Top AI Repos — open-source AI, indexed and scored

Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.

Intellindust-AI-Lab/EdgeCrafter

Pytorch implementation of "EdgeCrafter: Compact ViTs for Edge Dense Prediction via Task-Specialized Distillation"

74Apache-2.0Python

View on GitHub Website

Stars: 291
Forks: 36
Watchers: 3
Open issues: 7
Contributors: 4

Star history, last 90 days+3 over 5 snapshots

2026-07-24 · 2882026-08-06 · 291

Star history, last 90 days
Date	Stars
2026-07-24	288
2026-07-25	289
2026-07-28	291
2026-07-30	291
2026-08-06	291

Today

— stars today

This week

— stars this week

This month

— stars this month

Momentum

0.0

growth rate 0.00%/day

From the README

<h1 align="center">EdgeCrafter: Compact ViTs for Edge Dense Prediction via
Task-Specialized Distillation</h1>

<h3 align="center">
  <b>English</b> | <a href="README_zh.md">简体中文</a>
</h3>

<p align="center">
  <a href="https://intellindust-ai-lab.github.io/projects/EdgeCrafter/"><img src="https://img.shields.io/badge/Webpage-EdgeCrafter-blue.svg" alt="Webpage"></a>
  <a href="https://arxiv.org/abs/2603.18739"><img src="https://img.shields.io/badge/arXiv-EdgeCrafter-orange.svg" alt="arXiv"></a>
  <a href="#"><img src="https://img.shields.io/badge/License-Apache%202.0-green.svg" alt="License"></a>

</p>

<p align="center">
        <a href="https://capsule2077.github.io/">Longfei Liu <sup>*</sup><sup>‡</sup></a>&nbsp;
        <a >Yongjie Hou <sup>*</sup></a>&nbsp;
        <a >Yang Li <sup>*</sup></a>&nbsp;
        <a href='https://qiruiwang0728.github.io/homepage/'>Qirui Wang <sup>*</sup></a>&nbsp;
        <a >Youyang Sha</a>&nbsp; </br>
        <a >Yongjun Yu</a>&nbsp;
        <a >Yinzhi Wang</a>&nbsp;
        <a >Peizhe Ru</a>&nbsp;
        <a href="https://xuanlong-yu.github.io/">Xuanlong Yu<sup>†</sup></a>&nbsp
        <a href="https://xishen0220.github.io/">Xi Shen <sup>†</sup></a> <br><br>
      <a> * Equal Contribution &nbsp;&nbsp; ‡ Project Leader &nbsp;&nbsp; † Corresponding Author</a> <br>
</p>

<p align="center">
    <sup></sup> <a href="https://intellindust-ai-lab.github.io">Intellindust AI Lab</a> <br> 
</p>

<p align="center" style="margin:0; padding:0;">
  <img src=".github/teaser.png">
</p>


---

## 🚀 Updates
- **[2026-07-23]** EdgeCrafter has been integrated into [LightlyTrain](https://docs.lightly.ai/train/stable/pretrain_distill/models/edgecrafter.html)
- **[2026-07-22]** EdgeCrafter has been integrated into [LibreYOLO](https://www.libreyolo.com/zh).
- **[2026-04-20]** Our previous version [DEIMv2](https://github.com/Intellindust-AI-Lab/DEIMv2) was used by two winning teams at the [CVPR 2026 Maritime Computer Vision Workshop](https://arxiv.org/abs/2604.13244), taking home 2nd place in Thermal Object Detection Challenge and 3rd place in Vision-to-Chart Data Association Challenge.
- **[2026-03-21]** <a href="https://huggingface.co/Intellindust">We have uploaded our models on 🤗 Hugging Face</a>.
- **[2026-03-19]** Initial release of EdgeCrafter.

---

## 🤗 Hugging Face

<a href="https://huggingface.co/Intellindust">We have uploaded our models on 🤗 Hugging Face</a>! You can also access these models via [hf_models.ipynb](./hf_models.ipynb). Have a try!

---

## 📍 Reproducing the Results

- **Detection & Instance Segmentation:** [Instructions](./ecdetseg)
- **Pose Estimation:** [Instructions](./ecpose)

---

## 🏆 Model Zoo

### COCO2017 Validation Results

> **Note**: Latency is measured on an NVIDIA T4 GPU with batch size 1 under FP16 precision using TensorRT (v10.6).

### Object Detection

| Model | Size | AP<sub>50:95</sub> | #Params | GFLOPs | Latency (ms) | Config | Log | Checkpoint |
|:-----:|:----:|:--:|:-------:|:------:|:------------:|:------:|:---:|:----------:|
| **ECDet-S** | 640 | 51.7 | 10 | 26 | 5.41 | [config](ecdetseg/configs/ecdet/ecdet_s.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_s.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_s.pth) |
| **ECDet-M** | 640 | 54.3 | 18 | 53 | 7.98 | [config](ecdetseg/configs/ecdet/ecdet_m.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_m.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_m.pth) |
| **ECDet-L** | 640 | 57.0 | 31 | 101 | 10.49 | [config](ecdetseg/configs/ecdet/ecdet_l.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_l.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_l.pth) |
| **ECDet-X** | 640 | 57.9 | 49 | 151 | 12.70 | [config](ecdetseg/configs/ecdet/ecdet_x.yml) | [log](https://github.com/

Excerpt of 8,559 characters

Read on GitHub

Top contributors

No contributor data yet — it is fetched during the contributor sync pass.

Quality score

Would you bet a product on this? Bounded 0–100 and slow moving.

74/10074

Maintenance: 90%; How recently the default branch was pushed to
Releases: 25%; Recency and cadence of tagged releases
Community: 46%; Contributor breadth and bus factor
Issue backlog: 85%; Open issues measured against the size of the audience
Documentation: 99%; README depth, plus a homepage and topics
Licensing: 100%; How freely the code can be adopted

What to know before adopting

No tagged releases
Usable in commercial products with attribution.

Details

Created: 5 months ago
Last push: 14 days ago
Last release: none
Release tag: —
Releases in the last year: 0
Default branch: main
Size: 1.5k KB
Owner type: Organization
Owner location: China
Discovered via: topic:computer-vision
First indexed: 14 days ago
Last synced: 9 days ago

Intellindust-AI-Lab/EdgeCrafter

Pytorch implementation of "EdgeCrafter: Compact ViTs for Edge Dense Prediction via Task-Specialized Distillation"

74Apache-2.0Python

View on GitHub Website

Stars: 291
Forks: 36
Watchers: 3
Open issues: 7
Contributors: 4

Star history, last 90 days+3 over 5 snapshots

2026-07-24 · 2882026-08-06 · 291

Star history, last 90 days
Date	Stars
2026-07-24	288
2026-07-25	289
2026-07-28	291
2026-07-30	291
2026-08-06	291

Today

— stars today

This week

— stars this week

This month

— stars this month

Momentum

0.0

growth rate 0.00%/day

From the README

<h1 align="center">EdgeCrafter: Compact ViTs for Edge Dense Prediction via
Task-Specialized Distillation</h1>

<h3 align="center">
  <b>English</b> | <a href="README_zh.md">简体中文</a>
</h3>

<p align="center">
  <a href="https://intellindust-ai-lab.github.io/projects/EdgeCrafter/"><img src="https://img.shields.io/badge/Webpage-EdgeCrafter-blue.svg" alt="Webpage"></a>
  <a href="https://arxiv.org/abs/2603.18739"><img src="https://img.shields.io/badge/arXiv-EdgeCrafter-orange.svg" alt="arXiv"></a>
  <a href="#"><img src="https://img.shields.io/badge/License-Apache%202.0-green.svg" alt="License"></a>

</p>

<p align="center">
        <a href="https://capsule2077.github.io/">Longfei Liu <sup>*</sup><sup>‡</sup></a>&nbsp;
        <a >Yongjie Hou <sup>*</sup></a>&nbsp;
        <a >Yang Li <sup>*</sup></a>&nbsp;
        <a href='https://qiruiwang0728.github.io/homepage/'>Qirui Wang <sup>*</sup></a>&nbsp;
        <a >Youyang Sha</a>&nbsp; </br>
        <a >Yongjun Yu</a>&nbsp;
        <a >Yinzhi Wang</a>&nbsp;
        <a >Peizhe Ru</a>&nbsp;
        <a href="https://xuanlong-yu.github.io/">Xuanlong Yu<sup>†</sup></a>&nbsp
        <a href="https://xishen0220.github.io/">Xi Shen <sup>†</sup></a> <br><br>
      <a> * Equal Contribution &nbsp;&nbsp; ‡ Project Leader &nbsp;&nbsp; † Corresponding Author</a> <br>
</p>

<p align="center">
    <sup></sup> <a href="https://intellindust-ai-lab.github.io">Intellindust AI Lab</a> <br> 
</p>

<p align="center" style="margin:0; padding:0;">
  <img src=".github/teaser.png">
</p>


---

## 🚀 Updates
- **[2026-07-23]** EdgeCrafter has been integrated into [LightlyTrain](https://docs.lightly.ai/train/stable/pretrain_distill/models/edgecrafter.html)
- **[2026-07-22]** EdgeCrafter has been integrated into [LibreYOLO](https://www.libreyolo.com/zh).
- **[2026-04-20]** Our previous version [DEIMv2](https://github.com/Intellindust-AI-Lab/DEIMv2) was used by two winning teams at the [CVPR 2026 Maritime Computer Vision Workshop](https://arxiv.org/abs/2604.13244), taking home 2nd place in Thermal Object Detection Challenge and 3rd place in Vision-to-Chart Data Association Challenge.
- **[2026-03-21]** <a href="https://huggingface.co/Intellindust">We have uploaded our models on 🤗 Hugging Face</a>.
- **[2026-03-19]** Initial release of EdgeCrafter.

---

## 🤗 Hugging Face

<a href="https://huggingface.co/Intellindust">We have uploaded our models on 🤗 Hugging Face</a>! You can also access these models via [hf_models.ipynb](./hf_models.ipynb). Have a try!

---

## 📍 Reproducing the Results

- **Detection & Instance Segmentation:** [Instructions](./ecdetseg)
- **Pose Estimation:** [Instructions](./ecpose)

---

## 🏆 Model Zoo

### COCO2017 Validation Results

> **Note**: Latency is measured on an NVIDIA T4 GPU with batch size 1 under FP16 precision using TensorRT (v10.6).

### Object Detection

| Model | Size | AP<sub>50:95</sub> | #Params | GFLOPs | Latency (ms) | Config | Log | Checkpoint |
|:-----:|:----:|:--:|:-------:|:------:|:------------:|:------:|:---:|:----------:|
| **ECDet-S** | 640 | 51.7 | 10 | 26 | 5.41 | [config](ecdetseg/configs/ecdet/ecdet_s.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_s.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_s.pth) |
| **ECDet-M** | 640 | 54.3 | 18 | 53 | 7.98 | [config](ecdetseg/configs/ecdet/ecdet_m.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_m.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_m.pth) |
| **ECDet-L** | 640 | 57.0 | 31 | 101 | 10.49 | [config](ecdetseg/configs/ecdet/ecdet_l.yml) | [log](https://github.com/capsule2077/edgecrafter/raw/refs/heads/main/logs/ecdet_l.log) | [model](https://github.com/capsule2077/edgecrafter/releases/download/edgecrafterv1/ecdet_l.pth) |
| **ECDet-X** | 640 | 57.9 | 49 | 151 | 12.70 | [config](ecdetseg/configs/ecdet/ecdet_x.yml) | [log](https://github.com/

Excerpt of 8,559 characters

Read on GitHub

Top contributors

No contributor data yet — it is fetched during the contributor sync pass.

Quality score

Would you bet a product on this? Bounded 0–100 and slow moving.

74/10074

Maintenance: 90%; How recently the default branch was pushed to
Releases: 25%; Recency and cadence of tagged releases
Community: 46%; Contributor breadth and bus factor
Issue backlog: 85%; Open issues measured against the size of the audience
Documentation: 99%; README depth, plus a homepage and topics
Licensing: 100%; How freely the code can be adopted

What to know before adopting

No tagged releases
Usable in commercial products with attribution.

Details

Created: 5 months ago
Last push: 14 days ago
Last release: none
Release tag: —
Releases in the last year: 0
Default branch: main
Size: 1.5k KB
Owner type: Organization
Owner location: China
Discovered via: topic:computer-vision
First indexed: 14 days ago
Last synced: 9 days ago

Top AI Repos — open-source AI, indexed and scored

Intellindust-AI-Lab/EdgeCrafter

From the README

Top contributors

Quality score

What to know before adopting

Categories

Details

Intellindust-AI-Lab/EdgeCrafter

From the README

Top contributors

Quality score

What to know before adopting

Categories

Details

Top AI Repos — open-source AI, indexed and scored

Trend

From the README

Top contributors

Quality score

What to know before adopting

Categories

Details

Trend

From the README

Top contributors

Quality score

What to know before adopting

Categories

Details