Social Sciences DataLab has released Indra OCR, a new 1-million-parameter OCR model designed for high throughput and speed on CPUs. It can process up to 49 pages per second, significantly outperforming Tesseract and PaddleOCR in speed while also achieving competitive accuracy on benchmarks like OmniParseBench. Indra OCR outputs structured data including line and block bounding boxes with labels, and can handle digital PDFs by extracting existing text layers and only OCRing missing content. AI
IMPACT Offers a faster and more efficient OCR solution for document processing, potentially improving workflows for businesses and researchers.
RANK_REASON This is a new model release, but it is from a smaller lab and not a frontier model release.
Read on Hugging Face Trending Models →
- Apache Software License 2.0
- Chandra
- Hugging Face
- Indra OCR
- OmniParseBench
- ONNX Runtime
- OpenRAIL-M
- PaddleOCR
- Social Sciences DataLab
- Tesseract
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
