Bill of Lading OCR Benchmark
Models transcribe 25 synthetic bills of lading. A fixed Gemini model extracts 21 fields from each transcription for comparison with the known values.
The maker ranks 24 parsers, including 15 vision models. It publishes separate scores for digital documents, office scans, and simulated faxes.
- Made by
- okraPDF
- Released
- August 2026
- Kind
- Documents, charts and text in images
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
-
Gemini 3.7 Flash #26 on the capability index
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
okraPDF (/v1 production) tops the leaderboard with 96.9% Gemini 3.1 Flash-Lite leads with 94.9% among models released by August 2026
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 Gemini 3.1 Flash-Lite, at 94.9%. This view shows only models we track, each at its best.
The newest model with a score here came out in August 2026, 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
- 96.9 okraPDF (/v1 production)
- 96.6 Mistral OCR
- 94.9 Qwen3-VL 235B
- 94.9
Gemini 3.1 Flash-Lite
- 94.5 Claude Sonnet 5
- 93.5 Qwen3-VL 8B
- 92.1 LlamaParse
- 90.8 Reducto
- 89.9 GLM-4.6V
- 89.8
Gemini 3.7 Flash
- 89.3
Mistral Small 3
- 89.1
Llama 4 Maverick
- 88.9 Chandra (Datalab)
- 88.8 Kimi K2.5
- 88.0 GPT-5 mini
- 87.6
Gemini 2.5 Flash
- 86.4 Claude Haiku 4.5
- 85.1 GPT-5.4 Mini
- 84.1
Nova 2 Lite
- 83.3 GPT-4o mini
- 79.1 Tesseract 5
- 52.4 pdftotext (poppler)
- 52.4 PyMuPDF
- 52.1 pdfplumber
- 1 okraPDF (/v1 production) 96.9%
- 2 Mistral OCR 96.6%
- 3 Qwen3-VL 235B 94.9%
- 4
Gemini 3.1 Flash-Lite 94.9%
- 5 Claude Sonnet 5 94.5%
- 6 Qwen3-VL 8B 93.5%
- 7 LlamaParse 92.1%
- 8 Reducto 90.8%
- 9 GLM-4.6V 89.9%
- 10
Gemini 3.7 Flash 89.8%
- 11
Mistral Small 3 89.3%
- 12
Llama 4 Maverick 89.1%
- 13 Chandra (Datalab) 88.9%
- 14 Kimi K2.5 88.8%
- 15 GPT-5 mini 88%
- 16
Gemini 2.5 Flash Sep 2025 87.6%
- 17 Claude Haiku 4.5 86.4%
- 18 GPT-5.4 Mini 85.1%
- 19
Nova 2 Lite 84.1%
- 20 GPT-4o mini 83.3%
- 21 Tesseract 5 79.1%
- 22 pdftotext (poppler) 52.4%
- 23 PyMuPDF 52.4%
- 24 pdfplumber 52.1%
Source: Official Bill of Lading OCR Benchmark leaderboard
- 1
Gemini 3.1 Flash-Lite 94.9%
- 2 Claude Sonnet 5 94.5%
- 3 GLM-4.6V 89.9%
- 4
Gemini 3.7 Flash 89.8%
- 5
Mistral Small 3 89.3%
- 6
Llama 4 Maverick 89.1%
- 7 Kimi K2.5 88.8%
- 8 GPT-5 mini 88%
- 9
Gemini 2.5 Flash Sep 2025 87.6%
- 10 Claude Haiku 4.5 86.4%
- 11 GPT-5.4 Mini 85.1%
- 12
Nova 2 Lite 84.1%
- 13 GPT-4o mini 83.3%
Source: Official Bill of Lading OCR Benchmark leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | okraPDF (/v1 production) | 96.9% | Official leaderboard | ||
| 2 | Mistral OCR | 96.6% | Official leaderboard | ||
| 3 | Qwen3-VL 235B | 94.9% | Official leaderboard | ||
| 4 | | 94.9% | Mar 2026 | Official leaderboard | |
| 5 | Claude Sonnet 5 | 94.5% | Jun 2026 | Official leaderboard | |
| 6 | Qwen3-VL 8B | 93.5% | Official leaderboard | ||
| 7 | LlamaParse | 92.1% | Official leaderboard | ||
| 8 | Reducto | 90.8% | Official leaderboard | ||
| 9 | GLM-4.6V | 89.9% | Dec 2025 | Official leaderboard | |
| 10 | | 89.8% | Aug 2026 | Official leaderboard | |
| 11 | | 89.3% | Jan 2025 | Official leaderboard | |
| 12 | | 89.1% | Apr 2025 | Official leaderboard | |
| 13 | Chandra (Datalab) | 88.9% | Official leaderboard | ||
| 14 | Kimi K2.5 | 88.8% | Jan 2026 | Official leaderboard | |
| 15 | GPT-5 mini | 88% | Aug 2025 | Official leaderboard | |
| 16 | | 87.6% | Sep 2025 | Official leaderboard | |
| 17 | Claude Haiku 4.5 | 86.4% | Oct 2025 | Official leaderboard | |
| 18 | GPT-5.4 Mini | 85.1% | Mar 2026 | Official leaderboard | |
| 19 | | 84.1% | Dec 2025 | Official leaderboard | |
| 20 | GPT-4o mini | 83.3% | Jul 2024 | Official leaderboard | |
| 21 | Tesseract 5 | 79.1% | Official leaderboard | ||
| 22 | pdftotext (poppler) | 52.4% | Official leaderboard | ||
| 23 | PyMuPDF | 52.4% | Official leaderboard | ||
| 24 | pdfplumber | 52.1% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 94.9% | Mar 2026 | Official leaderboard | |
| 2 | Claude Sonnet 5 | 94.5% | Jun 2026 | Official leaderboard | |
| 3 | GLM-4.6V | 89.9% | Dec 2025 | Official leaderboard | |
| 4 | | 89.8% | Aug 2026 | Official leaderboard | |
| 5 | | 89.3% | Jan 2025 | Official leaderboard | |
| 6 | | 89.1% | Apr 2025 | Official leaderboard | |
| 7 | Kimi K2.5 | 88.8% | Jan 2026 | Official leaderboard | |
| 8 | GPT-5 mini | 88% | Aug 2025 | Official leaderboard | |
| 9 | | 87.6% | Sep 2025 | Official leaderboard | |
| 10 | Claude Haiku 4.5 | 86.4% | Oct 2025 | Official leaderboard | |
| 11 | GPT-5.4 Mini | 85.1% | Mar 2026 | Official leaderboard | |
| 12 | | 84.1% | Dec 2025 | Official leaderboard | |
| 13 | GPT-4o mini | 83.3% | Jul 2024 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Digital, Scan and Fax-hard. Each ranks models on one part of the test. GPT-5 mini leads 1 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Digital
13 models we track have a score here.
- 1 GPT-5 mini ↑7 100%
- 2 Claude Sonnet 5 100%
- 3
Gemini 3.7 Flash ↑1 100%
- 4 GPT-5.4 Mini ↑7 99.5%
- 5 GLM-4.6V ↓2 99%
- 6
Gemini 3.1 Flash-Lite ↓5 99%
- 7
Mistral Small 3 ↓2 98%
- 8
Llama 4 Maverick ↓2 96.5%
- 9
Nova 2 Lite ↑3 94%
- 10 Claude Haiku 4.5 93.3%
Scan
13 models we track have a score here.
- 1
Mistral Small 3 ↑4 100%
- 2
Llama 4 Maverick ↑4 100%
- 3 GPT-5 mini ↑5 100%
- 4 Kimi K2.5 ↑3 100%
- 5
Gemini 3.1 Flash-Lite ↓4 100%
- 6 GPT-5.4 Mini ↑5 100%
- 7 Claude Sonnet 5 ↓5 100%
- 8
Gemini 3.7 Flash ↓4 100%
- 9 Claude Haiku 4.5 ↑1 99.1%
- 10
Nova 2 Lite ↑2 97.5%
Fax-hard
13 models we track have a score here.
- 1
Gemini 3.1 Flash-Lite 77.2%
- 2
Gemini 2.5 Flash Sep 2025 ↑7 74.6%
- 3 Claude Sonnet 5 ↓1 72.4%
- 4 GLM-4.6V ↓1 68.9%
- 5 Kimi K2.5 ↑2 64.9%
- 6
Llama 4 Maverick 54.4%
- 7
Mistral Small 3 ↓2 51.3%
- 8 Claude Haiku 4.5 ↑2 50.4%
- 9 GPT-4o mini ↑4 48.7%
- 10
Gemini 3.7 Flash ↓6 48.7%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Bill of Lading OCR Benchmark leaderboard
Mistral Small 3 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 Bill of Lading OCR Benchmark 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
8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (24 entries).
- Pass Models scored 13 models we track, plus 11 other entries on its leaderboard.
- Pass Newest score The newest is from August 2026, for Gemini 3.7 Flash.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Gemini 3.7 Flash (#26).
- Note Headroom Saturated. The best is 94.9%, past 85% since January 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on okrapdf.com: 13 rows for models we track and 11 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.