Document AI Benchmark Arena
Models extract fields from business documents using a common schema and OCR text. The scorer normalizes numbers and dates and allows small spelling differences in names and addresses.
DocAI Fabric compares six general-purpose models with its own pipeline on 500 documents. The leaderboard reports field accuracy, classification accuracy, hallucination rate, latency, and cost.
- Made by
- DocAI Fabric
- Released
- February 2026
- Kind
- Documents, charts and text in images
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- Kept private
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
-
Gemini 2.5 Pro Jun 2025 #97 on the capability index
- Last model added
- June 2025
- Scores refreshed
- November 20th, 2026
Platform (orchestrated) tops the leaderboard with 95.4% GPT-4.1 leads with 92.6% among models released by June 2025
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. The best model we track is GPT-4.1, at 92.6%. This view shows only models we track, each at its best.
The newest model with a score here came out in June 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 Platform (orchestrated) 95.4%
- 2 GPT-4.1 92.6%
- 3 Claude Sonnet 4 91.4%
- 4 GPT-4o Mar 2025 91.3%
- 5
Gemini 2.5 Pro Jun 2025 90.2%
- 6
Qwen2.5-VL-72B 87.6%
- 7
Llama 4 Scout 85.4%
Source: Official Document AI Benchmark Arena leaderboard
- 1 GPT-4.1 92.6%
- 2 Claude Sonnet 4 91.4%
- 3 GPT-4o Mar 2025 91.3%
- 4
Gemini 2.5 Pro Jun 2025 90.2%
- 5
Qwen2.5-VL-72B 87.6%
- 6
Llama 4 Scout 85.4%
Source: Official Document AI Benchmark Arena leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Platform (orchestrated) | 95.4% | Official leaderboard | ||
| 2 | GPT-4.1 | 92.6% | Apr 2025 | Official leaderboard | |
| 3 | Claude Sonnet 4 | 91.4% | May 2025 | Official leaderboard | |
| 4 | GPT-4o Mar 2025 | 91.3% | Mar 2025 | Official leaderboard | |
| 5 | | 90.2% | Jun 2025 | Official leaderboard | |
| 6 | | 87.6% | Jan 2025 | Official leaderboard | |
| 7 | | 85.4% | Apr 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4.1 | 92.6% | Apr 2025 | Official leaderboard | |
| 2 | Claude Sonnet 4 | 91.4% | May 2025 | Official leaderboard | |
| 3 | GPT-4o Mar 2025 | 91.3% | Mar 2025 | Official leaderboard | |
| 4 | | 90.2% | Jun 2025 | Official leaderboard | |
| 5 | | 87.6% | Jan 2025 | Official leaderboard | |
| 6 | | 85.4% | Apr 2025 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Document AI Benchmark Arena leaderboard
Qwen2.5-VL-72B on the official leaderboard passed 85% in January 2025
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Document AI Benchmark Arena leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More 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 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access The tasks are private.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (7 entries).
- Pass Models scored 6 models we track, plus 1 other entry on its leaderboard.
- Note Newest score The newest came 17 months ago, in June 2025, for Gemini 2.5 Pro. That’s expected for a saturated test.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Gemini 2.5 Pro (#97).
- Note Headroom Saturated. The best is 92.6%, past 85% since January 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on docaifabric.com: 6 rows for models we track and 1 other entry 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.