Benchmark Almanac

Wild-OmniDocBench

In use

Models parse 1,350 photographed document pages with geometric distortion, lighting changes, and other capture artifacts. OmniDocBench metrics score text, formulas, tables, and reading order.

The data correspond to OmniDocBench v1.5. OpenBMB reports a deskewed-input track using the v1.6 scorer. That track must remain distinct from original-image results.

Released
March 2026
Kind
Documents, charts and text in images
Level
Expert
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images, Documents → Text, Tables
Interaction
Single turn
Run by
Anyone
In views
Vision
Last in a launch post
October 2026
Last model added
October 2026

In the authors’ words

Document parsing has recently advanced with multimodal large language models (MLLMs) that directly map document images to structured outputs. Traditional cascaded pipelines depend on precise layout analysis and often fail under casually captured or non-standard conditions. Although end-to-end approaches mitigate this dependency, they still exhibit repetitive, hallucinated, and structurally inconsistent predictions - primarily due to the scarcity of large-scale, high-quality full-page (document-level) end-to-end parsing data and the lack of structure-aware training strategies. To address these challenges, we propose a data-training co-design framework for robust end-to-end document parsing. A Realistic Scene Synthesis strategy constructs large-scale, structurally diverse full-page end-to-end supervision by composing layout templates with rich document elements, while a Document-Aware Training Recipe introduces progressive learning and structure-token optimization to enhance structural fidelity and decoding stability. We further build Wild-OmniDocBench, a benchmark derived from real-world captured documents for robustness evaluation. Integrated into a 1B-parameter MLLM, our method achieves superior accuracy and robustness across both scanned/digital and real-world captured scenarios. All models, data synthesis pipelines, and benchmarks will be publicly released to advance future research in document understanding.

Abstract of Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training (arXiv:2603.23885, CC0)
1
Model scored in lab cards
1
Launch post lists it, from 1 lab

MiniCPM-OCR-1.0 leads with 92.4%

Best score per model as labs report them in their own cards, drawn striped. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 1 MiniCPM-OCR-1.0 92.4%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 MiniCPM-OCR-1.0 92.4% Oct 2026 Lab model card

OpenBMB (ModelBest)’s own number for MiniCPM-OCR-1.0 passed 85% in October 2026

That was 7 months after its release. By the same age, 26% of the benchmarks from 2026 had passed 85%. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Reported by the lab in its model card
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) MiniCPM-OCR-1.0 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) MiniCPM-OCR-1.0

Source: Labs’ model cards

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

OpenBMB (ModelBest)
DateLaunch postScoreAs the post writes it
Oct 28, 2026 MiniCPM-OCR-1.0 92.4 Wild-OmniDocBench

More in Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

All 4 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 pass 4 notes
  • Note Independent runs Only labs’ own numbers, so no one else has checked them. Scores from launch posts (1 score).
  • Note Models scored Only 1 model we track.
  • Pass Newest score The newest is from October 2026, for MiniCPM-OCR-1.0.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Saturated. The best is 92.4%, past 85% since October 2026.
Leaderboard import 1 note
  • Note Importer No public leaderboard to read.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check