Instructions to use shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b with Transformers:

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

pipe = pipeline("image-text-to-text", model="shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b")
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("shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b",
		"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/shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b

SGLang

How to use shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b 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 "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b" \
    --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": "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b",
		"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 "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b" \
        --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": "shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b",
		"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 shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b with Docker Model Runner:
```
docker model run hf.co/shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b
```

pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b Model Card

Note: This is the pretrained model used for OLA-VLM-CLIP-ConvNeXT-Llama3-8b.

OLA-VLM distills target visual information into the intermediate representations of the LLM from a set of target encoders. It adopts a predictive embedding optimization approach at selected LLM layers during training to minimize the embedding losses along with the next token prediction (NTP) objective, resulting in a vision-centric approach to training the Multimodal Large Language Model.

GitHub Repo: https://github.com/SHI-Labs/OLA-VLM
Project Page: https://praeclarumjj3.github.io/ola_vlm/

Citation

If you found our work useful in your research, please consider starring ⭐ us on GitHub and citing 📚 us in your research!

@article{jain2024ola_vlm,
    title={{OLA-VLM: Elevating Visual Perception in Multimodal LLMs with Auxiliary Embedding Distillation}},
    author={Jitesh Jain and Zhengyuan Yang and Humphrey Shi and Jianfeng Gao and Jianwei Yang},
    journal={arXiv},
    year={2024}
}

Downloads last month: 1

Safetensors

Model size

9B params

Tensor type

F32

BF16

Model tree for shi-labs/pretrain_dsg_OLA-VLM-CLIP-ConvNeXT-Llama3-8b

Base model

meta-llama/Meta-Llama-3-8B-Instruct

Finetuned

(1129)

this model

Finetunes

1 model