Instructions to use zenoda/trocr-captcha-killer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zenoda/trocr-captcha-killer with Transformers:

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

pipe = pipeline("image-text-to-text", model="zenoda/trocr-captcha-killer")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("zenoda/trocr-captcha-killer")
model = AutoModelForMultimodalLM.from_pretrained("zenoda/trocr-captcha-killer")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zenoda/trocr-captcha-killer with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/zenoda/trocr-captcha-killer

SGLang

How to use zenoda/trocr-captcha-killer 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 "zenoda/trocr-captcha-killer" \
    --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": "zenoda/trocr-captcha-killer",
		"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 "zenoda/trocr-captcha-killer" \
        --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": "zenoda/trocr-captcha-killer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use zenoda/trocr-captcha-killer with Docker Model Runner:
```
docker model run hf.co/zenoda/trocr-captcha-killer
```

accuracy: 0.937338

from transformers import VisionEncoderDecoderModel, TrOCRProcessor
from PIL import Image
import requests

processor = TrOCRProcessor.from_pretrained("zenoda/trocr-captcha-killer")
model = VisionEncoderDecoderModel.from_pretrained("zenoda/trocr-captcha-killer")
model.to('cuda')
url = 'https://huggingface.co/datasets/zenoda/trocr-captcha-killer/resolve/main/106-1688354008849.png'
image = Image.open(requests.get(url, stream=True).raw).convert("RGB")
generated_ids = model.generate(processor(image, return_tensors="pt").pixel_values.to('cuda'))
predictText = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(predictText)

Downloads last month: 7

Inference Providers NEW

Image-Text-to-Text

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

zenoda
/

trocr-captcha-killer

Dataset used to train zenoda/trocr-captcha-killer