Instructions to use EditScore/EditScore-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EditScore/EditScore-7B with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("/share/project/luoxin/huggingface/hub/models--Qwen--Qwen2.5-VL-7B-Instruct/snapshots/cc594898137f460bfe9f0759e9844b3ce807cfb5")
model = PeftModel.from_pretrained(base_model, "EditScore/EditScore-7B")

Transformers

How to use EditScore/EditScore-7B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="EditScore/EditScore-7B")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("EditScore/EditScore-7B", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use EditScore/EditScore-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EditScore/EditScore-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EditScore/EditScore-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/EditScore/EditScore-7B

SGLang

How to use EditScore/EditScore-7B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "EditScore/EditScore-7B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EditScore/EditScore-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "EditScore/EditScore-7B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EditScore/EditScore-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use EditScore/EditScore-7B with Docker Model Runner:
```
docker model run hf.co/EditScore/EditScore-7B
```

sienna223 commited on Sep 30, 2025

Commit

ab9018c

verified ·

1 Parent(s): c558183

Update README.md

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -9,7 +9,7 @@ tags:
 ---
 <p align="center">
-  <img src="https://github.com/VectorSpaceLab/EditScore/blob/main/assets/logo.png" width="65%">
 </p>
 <p align="center">
@@ -49,7 +49,7 @@ To overcome this barrier, we provide a systematic, two-part solution:
 - **A Powerful & Versatile Tool**: Guided by our benchmark, we developed the **EditScore** model series. Through meticulous data curation and an effective self-ensembling strategy, EditScore sets a new state of the art for open-source reward models, even surpassing the accuracy of leading proprietary VLMs.
 <p align="center">
-  <img src="https://github.com/VectorSpaceLab/EditScore/blob/main/assets/table_reward_model_results.png" width="95%">
   <br>
   <em>Benchmark results on EditReward-Bench.</em>
 </p>
@@ -62,7 +62,7 @@ We demonstrate the practical utility of EditScore through two key applications:
 This repository releases both the **EditScore** models and the **EditReward-Bench** dataset to facilitate future research in reward modeling, policy optimization, and AI-driven model improvement.
 <p align="center">
-  <img src="https://github.com/VectorSpaceLab/EditScore/blob/main/assets/figure_edit_results.png" width="95%">
   <br>
   <em>EditScore as a superior reward signal for image editing.</em>
 </p>

 ---
 <p align="center">
+  <img src="assets/logo.png" width="65%">
 </p>
 <p align="center">
 - **A Powerful & Versatile Tool**: Guided by our benchmark, we developed the **EditScore** model series. Through meticulous data curation and an effective self-ensembling strategy, EditScore sets a new state of the art for open-source reward models, even surpassing the accuracy of leading proprietary VLMs.
 <p align="center">
+  <img src="assets/table_reward_model_results.png" width="95%">
   <br>
   <em>Benchmark results on EditReward-Bench.</em>
 </p>
 This repository releases both the **EditScore** models and the **EditReward-Bench** dataset to facilitate future research in reward modeling, policy optimization, and AI-driven model improvement.
 <p align="center">
+  <img src="assets/figure_edit_results.png" width="95%">
   <br>
   <em>EditScore as a superior reward signal for image editing.</em>
 </p>