Tech Jan 30, 2026 4 min PaddleOCR-VL-1.5 - document parsing SOTA with only 0.9B parameters Baidu's PaddleOCR-VL-1.5 reaches 94.5% accuracy on OmniDocBench v1.5 with just 0.9B parameters, surpassing large models such as GPT-4o and Qwen2.5-VL-72B. AI OCR VLM PaddlePaddle
