---
license: apache-2.0
base_model:
- ByteDance-Seed/BAGEL-7B-MoT
pipeline_tag: any-to-any
library_name: bagel-mot
arxiv: 2507.14119
---


# 🥯 BAGEL-NHR-Edit

<p align="left">
  <a href="https://riko0.github.io/No-Humans-Required/"> 🌐 NHR Website </a> | 
  <a href="https://arxiv.org/abs/2507.14119"> 📜 NHR Paper on arXiv </a> | 
  <a href="https://huggingface.co/datasets/iitolstykh/NHR-Edit"> 🤗 NHR-Edit Dataset (part 1) </a> | 
  <a href="https://huggingface.co/datasets/iitolstykh/NHR-Edit-part2"> 🤗 NHR-Edit Dataset (part 2) </a> | 
</p>

This repository hosts the model weights for **BAGEL**, fine-tuned on the **[NHR-Edit](https://huggingface.co/datasets/iitolstykh/NHR-Edit)** dataset (**on the part 1 only**). For installation, usage instructions, and further documentation, please visit the [official BAGEL GitHub repository](https://github.com/bytedance-seed/BAGEL).


### 🛠️ Training Setup

We performed parameter-efficient adaptation on the generation expert’s attention and FFN projection layers using LoRA.

LoRA parameters: 
```
r = 16
lora_alpha = 16
dropout = 0.05
bias = "none"
target_modules = [
  "v_proj_moe_gen",
  "k_proj_moe_gen",
  "mlp_moe_gen.down_proj",
  "mlp_moe_gen.gate_proj",
  "q_proj_moe_gen",
  "mlp_moe_gen.up_proj",
  "o_proj_moe_gen"
]
```

### 📊 Image Editing Metrics

#### Metrics for GEdit-Bench-EN:

| Model         | GEdit-Bench-EN (SC) ↑ | GEdit-Bench-EN (PQ) ↑ | GEdit-Bench-EN (O) ↑|
| ------------- | --------------------- | --------------------- | ------------------- |
| BAGEL-7B-MoT  |          7.983        |        6.570          |       6.921         |
| **BAGEL-NHR-Edit** | 8.067     | 6.881                 | 7.115               |
> *Scoring model:* `gpt-4.1-2025-04-14` *(with default temperature)*

#### Metrics for ImgEdit-Bench:

| Model         | Style | Extract | Remove | Background | Action | Adjust | Add | Replace | Compose | Overall ↑ |
| ------------- | ----- | ------- | ------ | -----------| ------ | ------ | ----| ------- | ------- | ------- |
| BAGEL-7B-MoT  |      4.22|        1.53|       3.04|      3.3|        4.07|       3.67|      3.98|       3.5 |       3.0 |       3.3 |
| **BAGEL-NHR-Edit** |      4.3|        1.62|       3.18|      3.42|        3.95|      3.55|      4.19|        3.77|       2.94|      3.39|
> *Scoring model:* `gpt-4o-2024-11-20` *(with temperature = 0.0)*

### 🖼️ Image Editing Results

Generated images for ImgEdit-Bench and GEdit-Bench are included in this repository.

Results comparison between original Bagel-7B-MoT and BAGEL-NHR-EDIT on samples from ImgEdit and GEdit benches:
![img](https://raw.githubusercontent.com/Riko0/No-Humans-Required-Dataset/refs/heads/main/images/Bagel_NHR_Edit_comp.jpg)

### License
**BAGEL-NHR-Edit** is licensed under the Apache 2.0 license. It is finetuned from [ByteDance-Seed/BAGEL-7B-MoT](https://huggingface.co/ByteDance-Seed/BAGEL-7B-MoT), which is also licensed under Apache 2.0.


### ✍️ Citation

```bibtex
@article{Layer2025NoHumansRequired,
    arxivId = {2507.14119},
    author = {Maksim Kuprashevich and Grigorii Alekseenko and Irina Tolstykh and Georgii Fedorov and Bulat Suleimanov and Vladimir Dokholyan and Aleksandr Gordeev},
    title = {{NoHumansRequired: Autonomous High-Quality Image Editing Triplet Mining}},
    year = {2025},
    eprint = {2507.14119},
    archivePrefix = {arXiv},
    primaryClass = {cs.CV},
    url = {https://arxiv.org/abs/2507.14119},
    journal={arXiv preprint arXiv:2507.14119}
}
```