ParseBench Leaderboard

Open paper
Benchmark Model Rank Results
document-parsing-on-parsebenchAnthropic Opus 5.5 (Effort High)#4Overall (ParseBench): 79.85Charts: 70.89Tables: 94.25…
document-parsing-on-parsebenchAnthropic Opus 5.5#5Overall (ParseBench): 78.01Charts: 64.12Tables: 93.86…
document-parsing-on-parsebenchOpenAI GPT-5.6 Sol (Reasoning High)#6Overall (ParseBench): 75.35Charts: 68.68Tables: 91.39…
document-parsing-on-parsebenchLightOnOCR-3-4B#7Overall (ParseBench): 75.1Charts: 66.1Tables: 83.8…
document-parsing-on-parsebenchLightOnOCR-3-0.8B#8Overall (ParseBench): 74.6Charts: 64.7Tables: 84.5…
document-parsing-on-parsebenchAnthropic Opus 5.5 (Effort Low)#9Overall (ParseBench): 74.47Charts: 53.52Tables: 92.69…
document-parsing-on-parsebenchInfinity-Parser2-Pro#10Overall (ParseBench): 74.28Charts: 61.3Tables: 86.4…
document-parsing-on-parsebenchInfinity-Parser2-Flash#11Overall (ParseBench): 73.25Charts: 55.56Tables: 82.88…
document-parsing-on-parsebenchMinerU2.5-Pro-2605-1.2B#12Overall (ParseBench): 72.78Charts: 61.64Tables: 77.59…
document-parsing-on-parsebenchGoogle Gemini 3.8 Flash (Thinking High)#13Overall (ParseBench): 72.08Charts: 32.34Tables: 89.12…
document-parsing-on-parsebenchLightOnOCR-3-1B#14Overall (ParseBench): 71.4Charts: 57.3Tables: 84.8…
document-parsing-on-parsebenchGoogle Gemini 3.7 Flash (Thinking High)#15Overall (ParseBench): 71.27Charts: 32.07Tables: 90.16…
document-parsing-on-parsebenchQwen 3.8 27B (Thinking)#16Overall (ParseBench): 70.79Charts: 69.17Tables: 66.82…
document-parsing-on-parsebenchAnthropic Fable 5#17Overall (ParseBench): 70.78Charts: 52.21Tables: 89.79…
document-parsing-on-parsebenchGoogle Gemini 3.7 Flash (Thinking Medium)#18Overall (ParseBench): 70.78Charts: 35.11Tables: 88.66…
document-parsing-on-parsebenchGLM-5.3 Flash#19Overall (ParseBench): 70.75Charts: 54.96Tables: 89.18…
document-parsing-on-parsebenchGoogle Gemini 3.8 Flash (Thinking Medium)#20Overall (ParseBench): 70.7Charts: 32.72Tables: 89.08…
document-parsing-on-parsebenchOpenAI GPT-6 Astra (Reasoning Low)#21Overall (ParseBench): 70.67Charts: 35.08Tables: 93.17…
document-parsing-on-parsebenchAnthropic Sonnet 5.5#22Overall (ParseBench): 70.18Charts: 37.69Tables: 91.04…
document-parsing-on-parsebenchGoogle Gemini 3.8 Flash (Thinking Low)#23Overall (ParseBench): 70.15Charts: 35.09Tables: 88.18…
document-parsing-on-parsebenchChandra-ocr-2#24Overall (ParseBench): 70.1Charts: 65.1Tables: 89.2…
document-parsing-on-parsebenchGoogle Gemini 3.7 Flash (Thinking Low)#25Overall (ParseBench): 69.57Charts: 31.84Tables: 87.96…
document-parsing-on-parsebenchOpenAI GPT-5.6 Terra (Reasoning Medium)#26Overall (ParseBench): 68.39Charts: 67.86Tables: 86.72…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning Max)#27Overall (ParseBench): 68.34Charts: 49.16Tables: 88.27…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning XHigh)#28Overall (ParseBench): 68.24Charts: 49.44Tables: 89.42…
document-parsing-on-parsebenchMistral OCR 4 (Annotation)#29Overall (ParseBench): 68.23Charts: 40.11Tables: 73.94…
document-parsing-on-parsebenchOpenAI GPT-5.6 Terra (Reasoning Low)#30Overall (ParseBench): 68.07Charts: 67.71Tables: 86.91…
document-parsing-on-parsebenchPaddleOCR-VL-1.6#31Overall (ParseBench): 67.43Charts: 54.24Tables: 67.77…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning High)#32Overall (ParseBench): 67.37Charts: 49.25Tables: 88.75…
document-parsing-on-parsebenchPaddleOCR-VL-1.5#33Overall (ParseBench): 65.95Charts: 47.62Tables: 67.38…
document-parsing-on-parsebenchDeepSeek V4.1 Flash (Thinking High)#34Overall (ParseBench): 65.72Charts: 51.11Tables: 82.82…
document-parsing-on-parsebenchSurya OCR 2#35Overall (ParseBench): 64.83Charts: 21.95Tables: 82.68…
document-parsing-on-parsebenchOpenAI GPT-5.6 Terra (Reasoning None)#36Overall (ParseBench): 64.17Charts: 63.75Tables: 86.23…
document-parsing-on-parsebenchHunyuanOCR 1.5#37Overall (ParseBench): 63.86Charts: 55.98Tables: 65.45…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning Medium)#38Overall (ParseBench): 63.58Charts: 46.91Tables: 85.32…
document-parsing-on-parsebenchGemma-4-31B-it#39Overall (ParseBench): 62.4Charts: 15.0Tables: 80.6…
document-parsing-on-parsebenchOpenAI GPT-5.6 Sol (Reasoning None)#40Overall (ParseBench): 62.12Charts: 59.57Tables: 89.34…
document-parsing-on-parsebenchQwen3-VL-8B-Instruct#41Overall (ParseBench): 61.97Charts: 28.18Tables: 74.61…
document-parsing-on-parsebenchQwen 3.8 27B#42Overall (ParseBench): 61.21Charts: 16.59Tables: 84.52…
document-parsing-on-parsebenchMistral OCR 4#43Overall (ParseBench): 60.68Charts: 2.35Tables: 73.94…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning Low)#44Overall (ParseBench): 60.18Charts: 38.95Tables: 81.67…
document-parsing-on-parsebenchTeleOCR#45Overall (ParseBench): 58.67Charts: 9.97Tables: 72.74…
document-parsing-on-parsebenchGemma-4-26B-A4B-it#46Overall (ParseBench): 58.5Charts: 14.2Tables: 70.0…
document-parsing-on-parsebenchDeepSeek V4.1 Flash (No Thinking)#47Overall (ParseBench): 56.57Charts: 20.46Tables: 79.65…
document-parsing-on-parsebenchOpenAI GPT-5.6 Luna (Reasoning None)#48Overall (ParseBench): 56.32Charts: 28.32Tables: 81.3…
document-parsing-on-parsebenchDots.mocr#49Overall (ParseBench): 55.79Charts: 0.95Tables: 85.15…
document-parsing-on-parsebenchHPD-Parsing#50Overall (ParseBench): 54.04Charts: 0.79Tables: 73.77…
document-parsing-on-parsebenchFalcon-OCR#51Overall (ParseBench): 53.08Charts: 0.82Tables: 74.7…
document-parsing-on-parsebenchDocling-models#52Overall (ParseBench): 50.65Charts: 52.76Tables: 66.41…
document-parsing-on-parsebenchNemotron-3-Nano-Omni-30B-A3B (Thinking)#53Overall (ParseBench): 48.51Charts: 31.47Tables: 70.44…
document-parsing-on-parsebenchQianfan-OCR#54Overall (ParseBench): 46.2Charts: 0.9Tables: 72.3…
document-parsing-on-parsebenchUnlimited-OCR#55Overall (ParseBench): 46.17Charts: 1.34Tables: 70.21…
document-parsing-on-parsebenchMinerU2.5-2509-1.2B#56Overall (ParseBench): 45.9Charts: 1.1Tables: 71.7…
document-parsing-on-parsebenchQwen3.6-35B-A3B#57Overall (ParseBench): 44.1Charts: 5.1Tables: 19.1…
document-parsing-on-parsebenchMinerU-Diffusion-V1-0320-2.5B#58Overall (ParseBench): 42.63Charts: 1.19Tables: 66.51…
document-parsing-on-parsebenchDeepSeek-OCR-2#59Overall (ParseBench): 41.2Charts: 1.1Tables: 61.7…
document-parsing-on-parsebenchPaddleOCR-VL#60Overall (ParseBench): 40.9Charts: 0.9Tables: 67.1…
document-parsing-on-parsebenchGemma-4-E4B-it#61Overall (ParseBench): 40.5Charts: 7.7Tables: 25.0…
document-parsing-on-parsebenchWeVisDoc-2B#62Overall (ParseBench): 38.93Charts: 0.86Tables: 70.41…
document-parsing-on-parsebenchWeVisDoc-4B#63Overall (ParseBench): 36.15Charts: 0.79Tables: 65.86…
document-parsing-on-parsebenchQwen3.5-4B#64Overall (ParseBench): 35.4Charts: 2.5Tables: 8.0…
document-parsing-on-parsebenchQwen3.5-9B#65Overall (ParseBench): 31.9Charts: 5.4Tables: 9.9…
document-parsing-on-parsebenchGLM-OCR#66Overall (ParseBench): 29.6Charts: 1.7Tables: 66.1…
document-parsing-on-parsebenchQwen3.5-0.8B#67Overall (ParseBench): 28.4Charts: 0.4Tables: 1.5…
document-parsing-on-parsebenchQwen3.5-2B#68Overall (ParseBench): 27.3Charts: 0.1Tables: 0.0…
document-parsing-on-parsebenchAmazon Nova 2 Lite–Overall (ParseBench): 40.43Charts: 6.7Tables: 68.38…
document-parsing-on-parsebenchAnthropic Fable 5.1–Overall (ParseBench): 78.92Charts: 67.06Tables: 91.52…
document-parsing-on-parsebenchAnthropic Haiku 4.5 (Disable Thinking)–Overall (ParseBench): 45.17Charts: 13.77Tables: 77.21…
document-parsing-on-parsebenchAnthropic Haiku 4.5 (Thinking)–Overall (ParseBench): 53.12Charts: 27.39Tables: 78.7…
document-parsing-on-parsebenchAnthropic Haiku 5.5–Overall (ParseBench): 57.4Charts: 10.13Tables: 86.85…
document-parsing-on-parsebenchAnthropic Opus 4.6–Overall (ParseBench): 54.07Charts: 13.49Tables: 86.52…
document-parsing-on-parsebenchAnthropic Opus 4.7–Overall (ParseBench): 63.34Charts: 55.84Tables: 87.17…
document-parsing-on-parsebenchAnthropic Opus 4.8–Overall (ParseBench): 63.7Charts: 49.75Tables: 89.65…
document-parsing-on-parsebenchAnthropic Sonnet 5–Overall (ParseBench): 62.13Charts: 60.53Tables: 86.65…
document-parsing-on-parsebenchGoogle Gemini 3 Flash (Thinking High)–Overall (ParseBench): 75.05Charts: 64.79Tables: 91.5…
document-parsing-on-parsebenchGoogle Gemini 3 Flash (Thinking Medium)–Overall (ParseBench): 75.0Charts: 61.56Tables: 91.01…
document-parsing-on-parsebenchGoogle Gemini 3 Flash (Thinking Minimal)–Overall (ParseBench): 71.04Charts: 64.83Tables: 89.85…
document-parsing-on-parsebenchGoogle Gemini 3.1 Flash Lite–Overall (ParseBench): 58.32Charts: 9.92Tables: 85.48…
document-parsing-on-parsebenchGoogle Gemini 3.1 Flash Lite (Thinking High)–Overall (ParseBench): 57.29Charts: 30.89Tables: 76.76…
document-parsing-on-parsebenchGoogle Gemini 3.1 Flash Lite (Thinking Medium)–Overall (ParseBench): 58.13Charts: 14.35Tables: 83.43…
document-parsing-on-parsebenchGoogle Gemini 3.1 Pro–Overall (ParseBench): 69.14Charts: 41.13Tables: 91.0…
document-parsing-on-parsebenchGoogle Gemini 3.1 Pro (Thinking Low)–Overall (ParseBench): 60.82Charts: 11.43Tables: 90.29…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash (Thinking High)–Overall (ParseBench): 74.02Charts: 45.27Tables: 90.44…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash (Thinking Medium)–Overall (ParseBench): 69.92Charts: 44.01Tables: 91.12…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash (Thinking Minimal)–Overall (ParseBench): 63.09Charts: 18.74Tables: 86.24…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash Lite–Overall (ParseBench): 57.15Charts: 6.26Tables: 73.48…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash Lite (Thinking High)–Overall (ParseBench): 60.64Charts: 19.27Tables: 83.31…
document-parsing-on-parsebenchGoogle Gemini 3.5 Flash Lite (Thinking Medium)–Overall (ParseBench): 54.3Charts: 6.23Tables: 72.85…
document-parsing-on-parsebenchGoogle Gemini 3.6 Flash (Thinking High)–Overall (ParseBench): 70.03Charts: 34.53Tables: 89.3…
document-parsing-on-parsebenchGoogle Gemini 3.6 Flash (Thinking Medium)–Overall (ParseBench): 66.78Charts: 31.0Tables: 89.47…
document-parsing-on-parsebenchGoogle Gemini 3.6 Flash (Thinking Minimal)–Overall (ParseBench): 65.6Charts: 24.76Tables: 85.5…
document-parsing-on-parsebenchGranite Vision 4.1 4B–Overall (ParseBench): 39.45Charts: 47.47Tables: 63.81…
document-parsing-on-parsebenchJina-OCR-v1–Overall (ParseBench): 45.93Charts: 7.52Tables: 74.9…
document-parsing-on-parsebenchKDL-Frontier-Parser-nano–Overall (ParseBench): 72.38Charts: 58.61Tables: 85.73…
document-parsing-on-parsebenchLightOnOCR-2-1B–Overall (ParseBench): 48.0Charts: 13.5Tables: 75.5…
document-parsing-on-parsebenchOpenAI GPT-5 Mini (Reasoning Medium)–Overall (ParseBench): 51.52Charts: 38.96Tables: 74.6…
document-parsing-on-parsebenchOpenAI GPT-5 Mini (Reasoning Minimal)–Overall (ParseBench): 46.83Charts: 30.13Tables: 69.82…
document-parsing-on-parsebenchOpenAI GPT-5.4 (Reasoning None)–Overall (ParseBench): 62.23Charts: 65.22Tables: 83.89…
document-parsing-on-parsebenchOpenAI GPT-5.4 Nano–Overall (ParseBench): 43.35Charts: 17.57Tables: 60.16…
document-parsing-on-parsebenchOpenAI GPT-5.5 (Reasoning Medium)–Overall (ParseBench): 67.76Charts: 65.53Tables: 90.05…
document-parsing-on-parsebenchOpenAI GPT-5.5 (Reasoning None)–Overall (ParseBench): 64.39Charts: 59.11Tables: 89.31…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning High)–Overall (ParseBench): 64.21Charts: 44.28Tables: 87.97…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning Low)–Overall (ParseBench): 59.32Charts: 39.05Tables: 84.2…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning Max)–Overall (ParseBench): 65.77Charts: 45.71Tables: 87.42…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning Medium)–Overall (ParseBench): 62.38Charts: 42.97Tables: 86.74…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning None)–Overall (ParseBench): 52.93Charts: 15.48Tables: 81.42…
document-parsing-on-parsebenchOpenAI GPT-6 Luna (Reasoning XHigh)–Overall (ParseBench): 64.32Charts: 42.38Tables: 88.01…
document-parsing-on-parsebenchOpenAI GPT-6 Sol (Reasoning High)–Overall (ParseBench): 69.69Charts: 48.05Tables: 92.63…
document-parsing-on-parsebenchOpenAI GPT-6 Sol (Reasoning Medium)–Overall (ParseBench): 68.19Charts: 46.46Tables: 90.27…
document-parsing-on-parsebenchOpenAI GPT-6 Sol (Reasoning None)–Overall (ParseBench): 66.1Charts: 42.02Tables: 89.49…
document-parsing-on-parsebenchOpenAI GPT-6.1 Sol (Reasoning Low)–Overall (ParseBench): 70.48Charts: 37.16Tables: 92.44…
document-parsing-on-parsebenchOvisOCR2–Overall (ParseBench): 40.84Charts: 1.01Tables: 77.37…
document-parsing-on-parsebenchaeddix-alpine-ocr (KDL pipeline)–Overall (ParseBench): 76.77Charts: 68.78Tables: 85.89…
document-parsing-on-parsebenchflorin-parser-nano–Overall (ParseBench): 76.69Charts: 65.19Tables: 86.1…
document-parsing-on-parsebenchrakedoc-nano–Overall (ParseBench): 77.23Charts: 64.89Tables: 86.44…

Rank counts only results with a code link.