Document Parsing on Real5-OmniDocBench benchmark leaderboard · SOTA2 Research

Document Parsing on Real5-OmniDocBench

93.19Overall Quality ScorePaddleOCR-VL-1.6

Evaluation Results

Method	Links
PaddleOCR-VL-1.62026.06		93.19	94.74	92.48	92.78	93.28	92.66
PaddleOCR-VL-1.52026.06		92.05	93.43	91.25	91.76	92.16	91.66
GLM-OCR2026.06		90.32	92.67	90.68	91.75	91.12	85.39
Gemini-3 Pro2026.06		89.24	89.47	88.9	88.86	89.53	89.45
MinerU2.5-pro2026.06		88.96	92.11	88.72	91.29	91.42	81.26
Qwen3-VL-235B-A22B-Instruct2026.06		88.904	89.43	89.99	89.27	89.27	86.56
Gemini-2.5 Pro2026.06		88.21	89.25	87.63	87.11	87.97	89.07
Qwen2.5-VL-72B2026.06		86.92	86.19	87.77	86.48	87.25	86.9
dots.ocr2026.06		86.38	86.87	86.01	87.18	87.57	84.27
MinerU2.52026.06		85.61	90.06	83.76	89.41	89.57	75.24
PaddleOCR-VL2026.06		85.54	92.11	85.97	82.54	89.61	77.47
Nanonets-OCR-s2026.06		84.19	85.52	83.56	84.86	85.01	81.98
MonkeyOCR-pro-3B2026.06		79.49	86.94	78.9	82.44	84.71	64.47
GPT-5.22026.06		78.66	84.43	76.26	76.75	80.88	75
MonkeyOCR-3B2026.06		78.29	84.65	77.27	80.71	83.16	65.67
MonkeyOCR-pro-1.2B2026.06		77.15	84.64	76.59	80.24	82.11	62.18
MinerU2-VLM2026.06		76.95	83.6	73.73	78.77	80.51	68.16
Deepseek-OCR2026.06		73.99	86.17	67.2	75.31	78.1	63.01
Deepseek-OCR 22026.06		73.01	89.59	66.53	71.65	76.02	61.28
PP-StructureV32026.06		64.45	84.68	59.34	66.89	73.38	37.98
Dolphin2026.06		61.78	72.16	60.35	64.29	67.29	44.83
Dolphin-1.52026.06		61.48	83.39	50.5	69.76	75.61	28.16
Maker-1.8.22026.06		60.1	70.27	58.98	63.65	66.31	41.27