Instructions to use DeepGlint-AI/UniME-Phi3.5-V-4.2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DeepGlint-AI/UniME-Phi3.5-V-4.2B with Transformers:

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

pipe = pipeline("image-text-to-text", model="DeepGlint-AI/UniME-Phi3.5-V-4.2B", trust_remote_code=True)
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)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("DeepGlint-AI/UniME-Phi3.5-V-4.2B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DeepGlint-AI/UniME-Phi3.5-V-4.2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DeepGlint-AI/UniME-Phi3.5-V-4.2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DeepGlint-AI/UniME-Phi3.5-V-4.2B",
		"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/DeepGlint-AI/UniME-Phi3.5-V-4.2B

SGLang

How to use DeepGlint-AI/UniME-Phi3.5-V-4.2B 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 "DeepGlint-AI/UniME-Phi3.5-V-4.2B" \
    --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": "DeepGlint-AI/UniME-Phi3.5-V-4.2B",
		"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 "DeepGlint-AI/UniME-Phi3.5-V-4.2B" \
        --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": "DeepGlint-AI/UniME-Phi3.5-V-4.2B",
		"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 DeepGlint-AI/UniME-Phi3.5-V-4.2B with Docker Model Runner:
```
docker model run hf.co/DeepGlint-AI/UniME-Phi3.5-V-4.2B
```

Kaichengalex commited on May 6, 2025

Commit

3b8e978

verified ·

1 Parent(s): 7a1c52d

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -2

README.md CHANGED Viewed

@@ -27,13 +27,17 @@ Yingda Chen,</span>
 [🏡 Project Page](https://garygutc.github.io/UniME) |  [📄 Paper](https://arxiv.org/pdf/2504.17432) | [💻 Github](https://github.com/deepglint/UniME)
 <p align="center">
-    <img src="figures/fig1.png">
 </p>
 ## 💡 Highlights
 To enhance the MLLM's embedding capability, we propose textual discriminative knowledge distillation. The training process involves decoupling the MLLM's LLM component and processing text with the prompt "Summarize the above sentences in one word.", followed by aligning the student (MLLM) and teacher (NV-Embed V2) embeddings via KL divergence on batch-wise similarity distributions. **Notably, only the LLM component is fine-tuned during this process, while all other parameters remain frozen**.
 <p align="center">

 [🏡 Project Page](https://garygutc.github.io/UniME) |  [📄 Paper](https://arxiv.org/pdf/2504.17432) | [💻 Github](https://github.com/deepglint/UniME)
+UniME achieves the top ranking on the MMEB leaderboard training using a 336×336 image resolution.（The screenshot is captured at 08:00 UTC+8 on May 6, 2025.）
 <p align="center">
+    <img src="figures/MMEB.png">
 </p>
 ## 💡 Highlights
+<p align="center">
+    <img src="figures/fig1.png">
+</p>
 To enhance the MLLM's embedding capability, we propose textual discriminative knowledge distillation. The training process involves decoupling the MLLM's LLM component and processing text with the prompt "Summarize the above sentences in one word.", followed by aligning the student (MLLM) and teacher (NV-Embed V2) embeddings via KL divergence on batch-wise similarity distributions. **Notably, only the LLM component is fine-tuned during this process, while all other parameters remain frozen**.
 <p align="center">