Instructions to use BleachNick/MMICL-Instructblip-T5-xxl with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BleachNick/MMICL-Instructblip-T5-xxl with Transformers:

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

pipe = pipeline("image-text-to-text", model="BleachNick/MMICL-Instructblip-T5-xxl")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("BleachNick/MMICL-Instructblip-T5-xxl")
model = AutoModelForImageTextToText.from_pretrained("BleachNick/MMICL-Instructblip-T5-xxl")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use BleachNick/MMICL-Instructblip-T5-xxl with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BleachNick/MMICL-Instructblip-T5-xxl"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BleachNick/MMICL-Instructblip-T5-xxl",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/BleachNick/MMICL-Instructblip-T5-xxl

SGLang

How to use BleachNick/MMICL-Instructblip-T5-xxl 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 "BleachNick/MMICL-Instructblip-T5-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": "BleachNick/MMICL-Instructblip-T5-xxl",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "BleachNick/MMICL-Instructblip-T5-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": "BleachNick/MMICL-Instructblip-T5-xxl",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use BleachNick/MMICL-Instructblip-T5-xxl with Docker Model Runner:
```
docker model run hf.co/BleachNick/MMICL-Instructblip-T5-xxl
```

Missing files

by sirwart - opened Aug 29, 2023

Discussion

sirwart

Aug 29, 2023

Is this repo missing the data files? When trying to run the example I get an error looking for https://huggingface.co/BleachNick/MMICL-Instructblip-T5-xxl/resolve/main/config.json, and looking at the files there's nothing apart from the README

BleachNick

Owner Aug 30, 2023

Sorry, currently the release of my model and data is still in progress, and the maintenance of the repository is not yet complete.
For the config.json file, you can simply use the same file from instructnlip.
You can use the code from our repository,
and try to load the model using the following script:

from model.instructblip import InstructBlipConfig, InstructBlipModel, InstructBlipPreTrainedModel,InstructBlipForConditionalGeneration,InstructBlipProcessor
import transformers
from PIL import Image
import torch
model_ckpt="BleachNick/MMICL-Instructblip-T5-xxl"
config_ckpt = "Salesforce/instructblip-flan-t5-xxl"
config = InstructBlipConfig.from_pretrained(config_ckpt )
model = InstructBlipForConditionalGeneration.from_pretrained(
        model_ckpt,
        config=config).to('cuda:0')

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment