Image-Text-to-Text
Transformers
Safetensors
English
qwen3_vl
qwen3-vl
vision-language
multimodal
visual-emotion
affective-computing
emotional-intelligence
reasoning
conversational
Eval Results (legacy)
Instructions to use wudq/EmObserver with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use wudq/EmObserver with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="wudq/EmObserver") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("wudq/EmObserver") model = AutoModelForMultimodalLM.from_pretrained("wudq/EmObserver", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use wudq/EmObserver with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "wudq/EmObserver" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wudq/EmObserver", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/wudq/EmObserver
- SGLang
How to use wudq/EmObserver 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 "wudq/EmObserver" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wudq/EmObserver", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'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 "wudq/EmObserver" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wudq/EmObserver", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use wudq/EmObserver with Docker Model Runner:
docker model run hf.co/wudq/EmObserver
Add comprehensive EmObserver model card
Browse files
README.md
CHANGED
|
@@ -82,6 +82,7 @@ The model accepts one or more images with a natural-language question and produc
|
|
| 82 |
| Resource | Description | Link |
|
| 83 |
| --- | --- | --- |
|
| 84 |
| Paper | MVEI & EmObserver | [arXiv:2607.21061](https://arxiv.org/abs/2607.21061) |
|
|
|
|
| 85 |
| Official code | Inference, evaluation, and four-stage training | [wdqqdw/EmObserver](https://github.com/wdqqdw/EmObserver) |
|
| 86 |
| Expanded release | Benchmarks, images, predictions, and training data | [wudq/MVEI_PLUS](https://huggingface.co/datasets/wudq/MVEI_PLUS) |
|
| 87 |
| EmObserver results | Released predictions and metric summaries | [MVEI_PLUS/baselines/Qwen3-VL-EMOBSERVER](https://huggingface.co/datasets/wudq/MVEI_PLUS/tree/main/baselines/Qwen3-VL-EMOBSERVER) |
|
|
@@ -300,7 +301,7 @@ The checkpoint is released under the Apache 2.0 license, following its [Qwen3-VL
|
|
| 300 |
|
| 301 |
## 馃摑 Citation
|
| 302 |
|
| 303 |
-
If you use EmObserver, MVEI, or
|
| 304 |
|
| 305 |
```bibtex
|
| 306 |
@article{wu2026mvei_emobserver,
|
|
@@ -309,5 +310,12 @@ If you use EmObserver, MVEI, or the expanded release, please cite:
|
|
| 309 |
journal = {arXiv preprint arXiv:2607.21061},
|
| 310 |
year = {2026}
|
| 311 |
}
|
| 312 |
-
```
|
| 313 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
| Resource | Description | Link |
|
| 83 |
| --- | --- | --- |
|
| 84 |
| Paper | MVEI & EmObserver | [arXiv:2607.21061](https://arxiv.org/abs/2607.21061) |
|
| 85 |
+
| Original paper (ICLR 2026) | Customizing Visual Emotion Evaluation for MLLMs | [OpenReview](https://openreview.net/forum?id=dQTSXWqZws) 路 [arXiv:2509.21950](https://arxiv.org/abs/2509.21950) |
|
| 86 |
| Official code | Inference, evaluation, and four-stage training | [wdqqdw/EmObserver](https://github.com/wdqqdw/EmObserver) |
|
| 87 |
| Expanded release | Benchmarks, images, predictions, and training data | [wudq/MVEI_PLUS](https://huggingface.co/datasets/wudq/MVEI_PLUS) |
|
| 88 |
| EmObserver results | Released predictions and metric summaries | [MVEI_PLUS/baselines/Qwen3-VL-EMOBSERVER](https://huggingface.co/datasets/wudq/MVEI_PLUS/tree/main/baselines/Qwen3-VL-EMOBSERVER) |
|
|
|
|
| 301 |
|
| 302 |
## 馃摑 Citation
|
| 303 |
|
| 304 |
+
If you use EmObserver or the expanded training and evaluation release, please cite the EmObserver paper below. When using the MVEI benchmark, dataset, or Emotion Statement Judgement formulation, please also cite the original ICLR 2026 paper.
|
| 305 |
|
| 306 |
```bibtex
|
| 307 |
@article{wu2026mvei_emobserver,
|
|
|
|
| 310 |
journal = {arXiv preprint arXiv:2607.21061},
|
| 311 |
year = {2026}
|
| 312 |
}
|
|
|
|
| 313 |
|
| 314 |
+
@inproceedings{wu2026customizing,
|
| 315 |
+
title = {Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approach},
|
| 316 |
+
author = {Wu, Daiqing and Yang, Dongbao and Zhao, Sicheng and Ma, Can and Zhou, Yu},
|
| 317 |
+
booktitle = {International Conference on Learning Representations},
|
| 318 |
+
year = {2026},
|
| 319 |
+
url = {https://openreview.net/forum?id=dQTSXWqZws}
|
| 320 |
+
}
|
| 321 |
+
```
|