Instructions to use fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct with Transformers:

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

pipe = pipeline("image-text-to-text", model="fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct", 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 AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct",
		"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/fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct

SGLang

How to use fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct 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 "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct" \
    --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": "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct",
		"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 "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct" \
        --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": "fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct",
		"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 fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct with Docker Model Runner:
```
docker model run hf.co/fudan-generative-ai/Bard-VL-B8-Mask-4B-Distil-Instruct
```

Update README.md

by merryyuan - opened about 1 month ago

base: refs/heads/main

←

from: refs/pr/6

Discussion Files changed

-17

Files changed (1) hide show

README.md +1 -17

README.md CHANGED Viewed

@@ -74,14 +74,9 @@ Compared with a standard autoregressive VLM release style, Bard-VL emphasizes:
 ## 🧭 Method Structure
 <p align="center">
-  <img src="./model.PNG" alt="Bard-VL method overview" width="100%">
 </p>
-<p align="center">
-  <em>Pipeline, block-wise attention mask, and mixed-noise scheduler used by Bard-VL.</em>
-</p>
----
 ## 📊 Evaluation Results
@@ -129,15 +124,6 @@ accelerate==1.12.0
 deepspeed==0.17.0
 ```
-Recommended runtime settings in the local repository:
-```bash
-dtype = bfloat16
-attn_implementation = sdpa
-block_size = 8
-denoising_steps = 8
-```
 ---
 ## 🚀 Inference Example
@@ -213,8 +199,6 @@ response_ids = model.generate(
 print(processor.tokenizer.batch_decode(response_ids, skip_special_tokens=True)[0].strip())
 ```
-For video understanding, replace the image message with the video example in `inference.py`.
 ---
 ## 📚 Citation

 ## 🧭 Method Structure
 <p align="center">
+  <img src="./model.PNG" alt="Bard-VL method overview" width="80%">
 </p>
 ## 📊 Evaluation Results
 deepspeed==0.17.0
 ```
 ---
 ## 🚀 Inference Example
 print(processor.tokenizer.batch_decode(response_ids, skip_special_tokens=True)[0].strip())
 ```
 ---
 ## 📚 Citation