Image-Text-to-Text
Transformers
Safetensors
English
multimodal
vision-language
image-quality-assessment
aesthetics
spatial-aesthetics
interior-design
Instructions to use AliHome3D/SA-IQA-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AliHome3D/SA-IQA-model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="AliHome3D/SA-IQA-model")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AliHome3D/SA-IQA-model", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AliHome3D/SA-IQA-model with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AliHome3D/SA-IQA-model" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AliHome3D/SA-IQA-model", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AliHome3D/SA-IQA-model
- SGLang
How to use AliHome3D/SA-IQA-model 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 "AliHome3D/SA-IQA-model" \ --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": "AliHome3D/SA-IQA-model", "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 "AliHome3D/SA-IQA-model" \ --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": "AliHome3D/SA-IQA-model", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AliHome3D/SA-IQA-model with Docker Model Runner:
docker model run hf.co/AliHome3D/SA-IQA-model
Add paper and code links
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,23 +1,26 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
|
|
|
| 3 |
language:
|
| 4 |
-
|
| 5 |
library_name: transformers
|
|
|
|
| 6 |
pipeline_tag: image-text-to-text
|
| 7 |
-
base_model: AIDC-AI/Ovis2.5-9B
|
| 8 |
-
datasets:
|
| 9 |
-
- SA-BENCH
|
| 10 |
tags:
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
---
|
| 18 |
|
| 19 |
# SA-IQA Model
|
| 20 |
|
|
|
|
|
|
|
|
|
|
| 21 |
SA-IQA is a multimodal image quality assessment model released with **“Beyond Pixels: Benchmarking and Reward-Based Assessing Framework for Visual Spatial Aesthetics.”**
|
| 22 |
|
| 23 |
The released final checkpoint is **`sa-iqa-prompt4`**, a fine-tuned model based on **Ovis2.5-9B** for assessing interior-image spatial aesthetics.
|
|
@@ -129,4 +132,4 @@ If you use this model, please cite:
|
|
| 129 |
booktitle={CVPR 2025 Workshop},
|
| 130 |
year={2025}
|
| 131 |
}
|
| 132 |
-
```
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model: AIDC-AI/Ovis2.5-9B
|
| 3 |
+
datasets:
|
| 4 |
+
- SA-BENCH
|
| 5 |
language:
|
| 6 |
+
- en
|
| 7 |
library_name: transformers
|
| 8 |
+
license: apache-2.0
|
| 9 |
pipeline_tag: image-text-to-text
|
|
|
|
|
|
|
|
|
|
| 10 |
tags:
|
| 11 |
+
- multimodal
|
| 12 |
+
- vision-language
|
| 13 |
+
- image-quality-assessment
|
| 14 |
+
- aesthetics
|
| 15 |
+
- spatial-aesthetics
|
| 16 |
+
- interior-design
|
| 17 |
---
|
| 18 |
|
| 19 |
# SA-IQA Model
|
| 20 |
|
| 21 |
+
**Paper:** [Beyond Pixels: Benchmarking and Reward-Based Assessing Framework for Visual Spatial Aesthetics](https://huggingface.co/papers/2512.05098)
|
| 22 |
+
**Code:** [https://github.com/AlibabaResearch/SA-IQA](https://github.com/AlibabaResearch/SA-IQA)
|
| 23 |
+
|
| 24 |
SA-IQA is a multimodal image quality assessment model released with **“Beyond Pixels: Benchmarking and Reward-Based Assessing Framework for Visual Spatial Aesthetics.”**
|
| 25 |
|
| 26 |
The released final checkpoint is **`sa-iqa-prompt4`**, a fine-tuned model based on **Ovis2.5-9B** for assessing interior-image spatial aesthetics.
|
|
|
|
| 132 |
booktitle={CVPR 2025 Workshop},
|
| 133 |
year={2025}
|
| 134 |
}
|
| 135 |
+
```
|