Loading...
Loading...
GPU-optimized OCR using Surya. Use when: (1) Extracting text from images/screenshots, (2) Processing PDFs with embedded images, (3) Multi-language document OCR, (4) Layout analysis and table detection. Supports 90+ languages with 2x accuracy over Tesseract.
npx skill4agent add aktsmm/agent-skills ocr-super-surya| Feature | Description |
|---|---|
| Accuracy | 2x better than Tesseract (0.97 vs 0.88) |
| GPU | PyTorch-based, CUDA optimized |
| Languages | 90+ including CJK |
| Layout | Document layout, table recognition |
# 1. Check GPU
python -c "import torch; print(f'CUDA: {torch.cuda.is_available()}')"
# 2. Install (with CUDA if GPU available)
pip install surya-ocr
# If CUDA=False but you have GPU, reinstall PyTorch:
pip uninstall torch torchvision torchaudio -y
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121# CLI
python scripts/ocr_helper.py image.png
python scripts/ocr_helper.py document.pdf -l ja en -o result.txt
# Or use surya directly
surya_ocr image.png --output_dir ./resultsfrom PIL import Image
from surya.recognition import RecognitionPredictor
from surya.detection import DetectionPredictor
from surya.foundation import FoundationPredictor
image = Image.open("document.png")
foundation_predictor = FoundationPredictor()
recognition_predictor = RecognitionPredictor(foundation_predictor)
detection_predictor = DetectionPredictor()
predictions = recognition_predictor([image], det_predictor=detection_predictor)
for page in predictions:
for line in page.text_lines:
print(line.text)| Variable | Default | Description |
|---|---|---|
| 512 | Reduce for lower VRAM |
| 36 | Reduce if OOM |
export RECOGNITION_BATCH_SIZE=256
surya_ocr image.png| Script | Description |
|---|---|
| Helper with OOM auto-retry, batch support |