Browse and Concentrate: Comprehending Multimodal Content via prior-LLM Context Fusion
Paper โข 2402.12195 โข Published
How to use wangphoebe/Brote-IM-XXL with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("image-text-to-text", model="wangphoebe/Brote-IM-XXL") # Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText
processor = AutoProcessor.from_pretrained("wangphoebe/Brote-IM-XXL")
model = AutoModelForImageTextToText.from_pretrained("wangphoebe/Brote-IM-XXL")How to use wangphoebe/Brote-IM-XXL with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "wangphoebe/Brote-IM-XXL"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "wangphoebe/Brote-IM-XXL",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/wangphoebe/Brote-IM-XXL
How to use wangphoebe/Brote-IM-XXL with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "wangphoebe/Brote-IM-XXL" \
--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": "wangphoebe/Brote-IM-XXL",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "wangphoebe/Brote-IM-XXL" \
--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": "wangphoebe/Brote-IM-XXL",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use wangphoebe/Brote-IM-XXL with Docker Model Runner:
docker model run hf.co/wangphoebe/Brote-IM-XXL
Models for this github repo that focuses on the modality isolation issues (image-text isolation and interimage isolation).
Detailed instructions are coming soon.