Benchmark Almanac

OmniParseBench

Gemini 4 Argon 96.0 leads since October 2026
Gemini 4 Argon 96.0 leads since October 2026
In use

Document parsers convert single pages to text, and 16,288 binary unit tests on 2,937 pages check the output. The headline joins text, table and layout subscores.

The results table also lists cost per 1,000 pages and refusals. Code is Apache 2.0 and page data keeps each source's license. Datalab includes its own parsers.

Made by
Datalab
Released
October 2026
Kind
Documents, charts and text in images
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Documents → Text, Tables
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 4 Carbon #3 on the capability index
Last in a launch post
October 2026
Last model added
October 2026
Scores refreshed
November 20th, 2026
34
Entries on its own leaderboard, 12 models we track
1
Launch post lists it, from 1 lab

Gemini 4 Argon leads with 96.0 Gemini 4 Argon leads with 96.0

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

Gemini 4 Argon leads with 96.0, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.

  1. 1 Gemini 4 Argon 96.0
  2. 2 Gemini 4 Carbon 95.7
  3. 3 Datalab accurate 94.4
  4. 4 Datalab balanced 92.9
  5. 5 Gemini 3.8 Flash 92.5
  6. 6 GPT-6.1 Sol 91.5
  7. 7 Reducto 91.1
  8. 8 LlamaParse 90.8
  9. 9 chandra-ocr-2 (Datalab) 90.4
  10. 10 Claude Opus 5.5 90.0
  11. 11 Mistral OCR 89.9
  12. 12 Claude Sonnet 5.5 88.9
  13. 13 dots.mocr 88.2
  14. 14 GPT-5.6 Sol 87.9
  15. 15 Infinity-Parser2-Pro 87.4
  16. 16 MinerU2.5-Pro 87.4
  17. 17 Surya OCR 2.5 (Datalab) 86.3
  18. 18 OXR-1.0 85.6
  19. 19 OvisOCR3 84.5
  20. 20 MiniCPM-OCR-1.0 83.5
  21. 21 Azure 83.4
  22. 22 Extend 83.4
  23. 23 Surya OCR 2 (Datalab) 83.3
  24. 24 LightOnOCR-3-4B 83.1
  25. 25 LightOnOCR-3-0.8B 81.9
  26. 26 GLM-OCR 81.9
  27. 27 Unlimited OCR 2 81.5
  28. 28 PaddleOCR-VL-1.6 81.3
  29. 29 Nemotron 3.5 Parse 79.9
  30. 30 Unlimited-OCR 79.5
  31. 31 Infinity-Parser2-Flash 78.7
  32. 32 DeepSeek-OCR 2 75.4
  33. 33 Indra OCR (Datalab) 71.4
  34. 34 HunyuanOCR-1.5 68.8

Source: Official OmniParseBench leaderboard

#EntryScoreSettingDateSource
1 Gemini 4 Argon 96.0 Sep 2026 Official leaderboard
2 Gemini 4 Carbon 95.7 Oct 2026 Official leaderboard
3 Datalab accurate closed weights 94.4 Official leaderboard
4 Datalab balanced closed weights 92.9 Official leaderboard
5 Gemini 3.8 Flash 92.5 Sep 2026 Official leaderboard
6 GPT-6.1 Sol 91.5 Sep 2026 Official leaderboard
7 Reducto closed weights 91.1 Official leaderboard
8 LlamaParse closed weights 90.8 Official leaderboard
9 chandra-ocr-2 (Datalab) open weights, 4.5B 90.4 Official leaderboard
10 Claude Opus 5.5 90.0 Sep 2026 Official leaderboard
11 Mistral OCR closed weights 89.9 Official leaderboard
12 Claude Sonnet 5.5 88.9 Sep 2026 Official leaderboard
13 dots.mocr open weights, 3B 88.2 Official leaderboard
14 GPT-5.6 Sol 87.9 Jul 2026 Official leaderboard
15 Infinity-Parser2-Pro open weights, 35.1B 87.4 Official leaderboard
16 MinerU2.5-Pro open weights, 1.2B 87.4 Official leaderboard
17 Surya OCR 2.5 (Datalab) open weights, 0.7B 86.3 Official leaderboard
18 OXR-1.0 85.6 Official leaderboard
19 OvisOCR3 open weights, 0.9B 84.5 Official leaderboard
20 MiniCPM-OCR-1.0 83.5 Oct 2026 Official leaderboard
21 Azure closed weights 83.4 Official leaderboard
22 Extend closed weights 83.4 Official leaderboard
23 Surya OCR 2 (Datalab) open weights, 0.7B 83.3 Official leaderboard
24 LightOnOCR-3-4B open weights, 4.5B 83.1 Official leaderboard
25 LightOnOCR-3-0.8B open weights, 0.9B 81.9 Official leaderboard
26 GLM-OCR 81.9 Jan 2026 Official leaderboard
27 Unlimited OCR 2 81.5 Oct 2026 Official leaderboard
28 PaddleOCR-VL-1.6 open weights, 1B 81.3 Official leaderboard
29 Nemotron 3.5 Parse open weights, 0.6B 79.9 Official leaderboard
30 Unlimited-OCR open weights, 3.3B 79.5 Official leaderboard
31 Infinity-Parser2-Flash open weights, 2.2B 78.7 Official leaderboard
32 DeepSeek-OCR 2 75.4 Jan 2026 Official leaderboard
33 Indra OCR (Datalab) open weights, 0B 71.4 Official leaderboard
34 HunyuanOCR-1.5 open weights, 1B 68.8 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 4 Argon 96.0 Sep 2026 Official leaderboard
2 Gemini 4 Carbon 95.7 Oct 2026 Official leaderboard
3 Gemini 3.8 Flash 92.5 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 91.5 Sep 2026 Official leaderboard
5 Claude Opus 5.5 90.0 Sep 2026 Official leaderboard
6 Claude Sonnet 5.5 88.9 Sep 2026 Official leaderboard
7 GPT-5.6 Sol 87.9 Jul 2026 Official leaderboard
8 OXR-1.0 85.6 Official leaderboard
9 MiniCPM-OCR-1.0 83.5 Oct 2026 Official leaderboard
10 GLM-OCR 81.9 Jan 2026 Official leaderboard
11 Unlimited OCR 2 81.5 Oct 2026 Official leaderboard
12 DeepSeek-OCR 2 75.4 Jan 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 4 Argon 96.0 Sep 2026 Official leaderboard
2 Gemini 4 Carbon 95.7 Oct 2026 Official leaderboard
3 Gemini 3.8 Flash 92.5 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 91.5 Sep 2026 Official leaderboard
5 Claude Opus 5.5 90.0 Sep 2026 Official leaderboard
6 Claude Sonnet 5.5 88.9 Sep 2026 Official leaderboard
7 GPT-5.6 Sol 87.9 Jul 2026 Official leaderboard
8 Surya OCR 2.5 85.9 Oct 2026 Lab model card
9 OXR-1.0 85.6 Official leaderboard
10 MiniCPM-OCR-1.0 83.5 Oct 2026 Official leaderboard
11 GLM-OCR 81.9 Jan 2026 Official leaderboard
12 Unlimited OCR 2 81.5 Oct 2026 Official leaderboard
13 DeepSeek-OCR 2 75.4 Jan 2026 Official leaderboard

Labs report scores for 1 model in their cards. Switch the scores above to rank them together.

Rankings by part

We read 3 parts from the benchmark’s own leaderboard: Text, Tables and Layout. Each ranks models on one part of the test. Gemini 4 Argon leads 2 of them, and Gemini 4 Carbon leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Text

12 models we track have a score here.

  1. 1 Gemini 4 Carbon ↑1 94.7
  2. 2 Gemini 4 Argon ↓1 94.4
  3. 3 Claude Opus 5.5 ↑2 93.9
  4. 4 Claude Sonnet 5.5 ↑2 93.7
  5. 5 Gemini 3.8 Flash ↓2 92.0
  6. 6 GPT-6.1 Sol ↓2 89.8
  7. 7 GPT-5.6 Sol 86.0
  8. 8 OXR-1.0 77.0
  9. 9 MiniCPM-OCR-1.0 73.4
  10. 10 Unlimited OCR 2 ↑1 73.0

Number one over time Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Own leaderboard 1 leader

Source: Official OmniParseBench leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Reported by the lab in its model card
75.0 80.0 85.0 90.0 95.0 100.0 2026 2027 75.0 80.0 85.0 90.0 95.0 100.0 2026 2027

Sources: Official OmniParseBench leaderboard and 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.

Datalab
DateLaunch postScoreAs the post writes it
Oct 25, 2026 Surya OCR 2.5 85.9 85.9 on OmniParseBench

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 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (34 entries) and launch posts (1 score).
  • Pass Models scored 13 models we track, plus 22 other entries on its leaderboard.
  • Pass Newest score The newest is from October 2026, for Gemini 4 Carbon.
  • Pass Frontier models 5 of the top 10 have a score, led by Gemini 4 Carbon (#3).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 12 rows for models we track and 22 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
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