Instructions to use AnyaSchen/vit-rugpt3-medium-pushkin with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AnyaSchen/vit-rugpt3-medium-pushkin with Transformers:

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

pipe = pipeline("image-text-to-text", model="AnyaSchen/vit-rugpt3-medium-pushkin")

# Load model directly
from transformers import AutoTokenizer, AutoModelForImageTextToText

tokenizer = AutoTokenizer.from_pretrained("AnyaSchen/vit-rugpt3-medium-pushkin")
model = AutoModelForImageTextToText.from_pretrained("AnyaSchen/vit-rugpt3-medium-pushkin")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use AnyaSchen/vit-rugpt3-medium-pushkin with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AnyaSchen/vit-rugpt3-medium-pushkin"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AnyaSchen/vit-rugpt3-medium-pushkin",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AnyaSchen/vit-rugpt3-medium-pushkin

SGLang

How to use AnyaSchen/vit-rugpt3-medium-pushkin 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 "AnyaSchen/vit-rugpt3-medium-pushkin" \
    --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": "AnyaSchen/vit-rugpt3-medium-pushkin",
		"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 "AnyaSchen/vit-rugpt3-medium-pushkin" \
        --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": "AnyaSchen/vit-rugpt3-medium-pushkin",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AnyaSchen/vit-rugpt3-medium-pushkin with Docker Model Runner:
```
docker model run hf.co/AnyaSchen/vit-rugpt3-medium-pushkin
```

This repo contains model for generation poetry from image in style of Pushkin.The model is fune-tuned concatecation of two pre-trained models: google/vit-base-patch16-224 as encoder and AnyaSchen/rugpt3_pushkin as decoder.

To use this model you can do:

from PIL import Image
import requests
from transformers import AutoTokenizer, VisionEncoderDecoderModel, ViTImageProcessor

def generate_poetry(fine_tuned_model, image, tokenizer):
    pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values
    pixel_values = pixel_values.to(device)

    # Generate the poetry with the fine-tuned VisionEncoderDecoder model
    generated_tokens = fine_tuned_model.generate(
        pixel_values,
        max_length=300,
        num_beams=3,
        top_p=0.8,
        temperature=2.0,
        do_sample=True,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )

    # Decode the generated tokens
    generated_poetry = tokenizer.decode(generated_tokens[0], skip_special_tokens=True)
    return generated_poetry

path = 'AnyaSchen/vit-rugpt3-medium-pushkin'
fine_tuned_model = VisionEncoderDecoderModel.from_pretrained(path).to(device)
feature_extractor = ViTImageProcessor.from_pretrained(path)
tokenizer = AutoTokenizer.from_pretrained(path)

url = 'https://anandaindia.org/wp-content/uploads/2018/12/happy-man.jpg'
image = Image.open(requests.get(url, stream=True).raw)

generated_poetry = generate_poetry(fine_tuned_model, image, tokenizer)
print(generated_poetry)

Downloads last month: 6

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

AnyaSchen
/

vit-rugpt3-medium-pushkin

Dataset used to train AnyaSchen/vit-rugpt3-medium-pushkin