Qianfan-OCR: A Unified End-to-End Model for Document Intelligence
This work addresses the need for efficient and accurate document intelligence tools for users handling complex layouts, though it is incremental in improving layout grounding within an end-to-end framework.
The authors tackled the problem of unifying document parsing, layout analysis, and understanding in a single model, resulting in a 4B-parameter end-to-end vision-language model that achieves top rankings on benchmarks like OmniDocBench v1.5 (93.12) and OlmOCR Bench (79.8), surpassing models such as Gemini-3.1-Pro.
We present Qianfan-OCR, a 4B-parameter end-to-end vision-language model that unifies document parsing, layout analysis, and document understanding within a single architecture. It performs direct image-to-Markdown conversion and supports diverse prompt-driven tasks including table extraction, chart understanding, document QA, and key information extraction. To address the loss of explicit layout analysis in end-to-end OCR, we propose Layout-as-Thought, an optional thinking phase triggered by special think tokens that generates structured layout representations -- bounding boxes, element types, and reading order -- before producing final outputs, recovering layout grounding capabilities while improving accuracy on complex layouts. Qianfan-OCR ranks first among end-to-end models on OmniDocBench v1.5 (93.12) and OlmOCR Bench (79.8), achieves competitive results on OCRBench, CCOCR, DocVQA, and ChartQA against general VLMs of comparable scale, and attains the highest average score on public key information extraction benchmarks, surpassing Gemini-3.1-Pro, Seed-2.0, and Qwen3-VL-235B. The model is publicly accessible via the Baidu AI Cloud Qianfan platform.