IDP Core Bench
Models extract fields and tables, transcribe text, and answer questions about documents. The test covers about 2,000 invoices, receipts, forms, and handwritten documents.
The makers list 27 models on the four-task leaderboard. They publish an overall score and a score for each task.
- Made by
- Nanonets and IIT Indore
- Kind
- Documents, charts and text in images
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Tasks
- All public
- Measures
- Capability
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-5.4 #28 on the capability index
- Last model added
- March 2026
- Scores refreshed
- November 20th, 2026
Gemini 3.1 Pro leads with 89.6% among models released by March 2026 Gemini 3.1 Pro leads with 89.6% among models released by March 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. This view shows only models we track, each at its best.
The newest model with a score here came out in March 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
- 89.6
Gemini 3.1 Pro
- 84.4 GPT-5.4
- 81.8
Gemini 3 Pro
- 81.2 Claude Sonnet 4.6
- 81.1 Claude Opus 4.6
- 80.5
Gemini 3 Flash
- 80.2 Nanonets OCR-3
- 80.0 Qwen3-VL-235B
- 79.8
Qwen3-VL-Plus
- 77.4 GPT-5.2
- 76.2
Qwen3.5-9B
- 74.7 GPT-4.1
- 74.5
Qwen3.5-4B
- 73.8 Nanonets OCR2+
- 73.3 GPT-5 mini
- 72.9 Claude Haiku 4.5
- 71.7
Ministral 8B
- 68.5
Mistral Small 4
- 67.1
Qwen3.5-2B
- 65.8 GPT-5 nano
- 61.2
Qwen3.5-0.8B
- 59.0
Pixtral 12B
- 58.6 Llama-3.2-Vision-11B
- 55.0
Gemma 4 E4B
- 54.9 GLM-OCR
- 44.1
Gemma 4 E2B
- 0.0
Gemma 3 12B
- 1
Gemini 3.1 Pro 89.6%
- 2 GPT-5.4 84.4%
- 3
Gemini 3 Pro 81.8%
- 4 Claude Sonnet 4.6 81.2%
- 5 Claude Opus 4.6 81.1%
- 6
Gemini 3 Flash 80.5%
- 7 Nanonets OCR-3 80.2%
- 8 Qwen3-VL-235B 80%
- 9
Qwen3-VL-Plus 79.8%
- 10 GPT-5.2 77.4%
- 11
Qwen3.5-9B 76.2%
- 12 GPT-4.1 74.7%
- 13
Qwen3.5-4B 74.5%
- 14 Nanonets OCR2+ 73.8%
- 15 GPT-5 mini 73.3%
- 16 Claude Haiku 4.5 72.9%
- 17
Ministral 8B 71.7%
- 18
Mistral Small 4 68.5%
- 19
Qwen3.5-2B 67.1%
- 20 GPT-5 nano 65.8%
- 21
Qwen3.5-0.8B 61.2%
- 22
Pixtral 12B 59%
- 23 Llama-3.2-Vision-11B 58.6%
- 24
Gemma 4 E4B 55%
- 25 GLM-OCR 54.9%
- 26
Gemma 4 E2B 44.1%
- 27
Gemma 3 12B 0%
Source: Official IDP Core Bench leaderboard
- 89.6
Gemini 3.1 Pro
- 84.4 GPT-5.4
- 81.8
Gemini 3 Pro
- 81.2 Claude Sonnet 4.6
- 81.1 Claude Opus 4.6
- 80.5
Gemini 3 Flash
- 79.8
Qwen3-VL-Plus
- 77.4 GPT-5.2
- 76.2
Qwen3.5-9B
- 74.7 GPT-4.1
- 74.5
Qwen3.5-4B
- 73.3 GPT-5 mini
- 72.9 Claude Haiku 4.5
- 71.7
Ministral 8B
- 68.5
Mistral Small 4
- 67.1
Qwen3.5-2B
- 65.8 GPT-5 nano
- 61.2
Qwen3.5-0.8B
- 59.0
Pixtral 12B
- 55.0
Gemma 4 E4B
- 54.9 GLM-OCR
- 44.1
Gemma 4 E2B
- 0.0
Gemma 3 12B
- 1
Gemini 3.1 Pro 89.6%
- 2 GPT-5.4 84.4%
- 3
Gemini 3 Pro 81.8%
- 4 Claude Sonnet 4.6 81.2%
- 5 Claude Opus 4.6 81.1%
- 6
Gemini 3 Flash 80.5%
- 7
Qwen3-VL-Plus 79.8%
- 8 GPT-5.2 77.4%
- 9
Qwen3.5-9B 76.2%
- 10 GPT-4.1 74.7%
- 11
Qwen3.5-4B 74.5%
- 12 GPT-5 mini 73.3%
- 13 Claude Haiku 4.5 72.9%
- 14
Ministral 8B 71.7%
- 15
Mistral Small 4 68.5%
- 16
Qwen3.5-2B 67.1%
- 17 GPT-5 nano 65.8%
- 18
Qwen3.5-0.8B 61.2%
- 19
Pixtral 12B 59%
- 20
Gemma 4 E4B 55%
- 21 GLM-OCR 54.9%
- 22
Gemma 4 E2B 44.1%
- 23
Gemma 3 12B 0%
Source: Official IDP Core Bench leaderboard
| # | Entry | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 89.6% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 2 | GPT-5.4 | 84.4% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 3 | | 81.8% | Nov 2025 | idp-leaderboard.org | Official leaderboard | |
| 4 | Claude Sonnet 4.6 | 81.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 5 | Claude Opus 4.6 | 81.1% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 6 | | 80.5% | Dec 2025 | idp-leaderboard.org | Official leaderboard | |
| 7 | Nanonets OCR-3 closed weights | 80.2% | idp-leaderboard.org | Official leaderboard | ||
| 8 | Qwen3-VL-235B open weights | 80% | idp-leaderboard.org | Official leaderboard | ||
| 9 | | 79.8% | Sep 2025 | idp-leaderboard.org | Official leaderboard | |
| 10 | GPT-5.2 | 77.4% | Dec 2025 | idp-leaderboard.org | Official leaderboard | |
| 11 | | 76.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 12 | GPT-4.1 | 74.7% | Apr 2025 | idp-leaderboard.org | Official leaderboard | |
| 13 | | 74.5% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 14 | Nanonets OCR2+ closed weights | 73.8% | idp-leaderboard.org | Official leaderboard | ||
| 15 | GPT-5 mini | 73.3% | Aug 2025 | idp-leaderboard.org | Official leaderboard | |
| 16 | Claude Haiku 4.5 | 72.9% | Oct 2025 | idp-leaderboard.org | Official leaderboard | |
| 17 | | 71.7% | Oct 2024 | idp-leaderboard.org | Official leaderboard | |
| 18 | | 68.5% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 19 | | 67.1% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 20 | GPT-5 nano | 65.8% | Aug 2025 | idp-leaderboard.org | Official leaderboard | |
| 21 | | 61.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 22 | | 59% | Sep 2024 | idp-leaderboard.org | Official leaderboard | |
| 23 | Llama-3.2-Vision-11B open weights | 58.6% | idp-leaderboard.org | Official leaderboard | ||
| 24 | | 55% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 25 | GLM-OCR | 54.9% | Jan 2026 | idp-leaderboard.org | Official leaderboard | |
| 26 | | 44.1% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 27 | | 0% | Mar 2025 | idp-leaderboard.org | Official leaderboard |
| # | Model | Score | Setting | Date | Link | Source |
|---|---|---|---|---|---|---|
| 1 | | 89.6% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 2 | GPT-5.4 | 84.4% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 3 | | 81.8% | Nov 2025 | idp-leaderboard.org | Official leaderboard | |
| 4 | Claude Sonnet 4.6 | 81.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 5 | Claude Opus 4.6 | 81.1% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 6 | | 80.5% | Dec 2025 | idp-leaderboard.org | Official leaderboard | |
| 7 | | 79.8% | Sep 2025 | idp-leaderboard.org | Official leaderboard | |
| 8 | GPT-5.2 | 77.4% | Dec 2025 | idp-leaderboard.org | Official leaderboard | |
| 9 | | 76.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 10 | GPT-4.1 | 74.7% | Apr 2025 | idp-leaderboard.org | Official leaderboard | |
| 11 | | 74.5% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 12 | GPT-5 mini | 73.3% | Aug 2025 | idp-leaderboard.org | Official leaderboard | |
| 13 | Claude Haiku 4.5 | 72.9% | Oct 2025 | idp-leaderboard.org | Official leaderboard | |
| 14 | | 71.7% | Oct 2024 | idp-leaderboard.org | Official leaderboard | |
| 15 | | 68.5% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 16 | | 67.1% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 17 | GPT-5 nano | 65.8% | Aug 2025 | idp-leaderboard.org | Official leaderboard | |
| 18 | | 61.2% | Feb 2026 | idp-leaderboard.org | Official leaderboard | |
| 19 | | 59% | Sep 2024 | idp-leaderboard.org | Official leaderboard | |
| 20 | | 55% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 21 | GLM-OCR | 54.9% | Jan 2026 | idp-leaderboard.org | Official leaderboard | |
| 22 | | 44.1% | Mar 2026 | idp-leaderboard.org | Official leaderboard | |
| 23 | | 0% | Mar 2025 | idp-leaderboard.org | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as KIE, OCR and Table. Gemini 3.1 Pro leads 3 of them, and Gemini 3 Flash leads the other one. Arrows show how far a model moved from its overall rank among the models we track.
KIE
23 models we track have a score here.
- 1
Gemini 3 Flash ↑5 91.1%
- 2 Claude Opus 4.6 ↑3 89.8%
- 3 Claude Sonnet 4.6 ↑1 89.5%
- 4 GPT-5.2 ↑4 87.5%
- 5 GPT-4.1 ↑5 87.1%
- 6
Gemini 3.1 Pro ↓5 86.8%
- 7
Qwen3.5-9B ↑2 86.5%
- 8
Qwen3.5-4B ↑3 86%
- 9
Ministral 8B ↑5 85.7%
- 10 GPT-5 mini ↑2 85.7%
OCR
23 models we track have a score here.
- 1
Gemini 3.1 Pro 82.8%
- 2
Gemini 3 Pro ↑1 81.8%
- 3
Gemini 3 Flash ↑3 81.7%
- 4 GPT-4.1 ↑6 75.6%
- 5 Claude Opus 4.6 74%
- 6
Gemma 4 E4B ↑14 74%
- 7 Claude Sonnet 4.6 ↓3 73.7%
- 8 GPT-5 mini ↑4 73%
- 9 GPT-5.2 ↓1 72.8%
- 10
Gemma 4 E2B ↑12 72.3%
Table
23 models we track have a score here.
- 1
Gemini 3.1 Pro 96.4%
- 2 Claude Sonnet 4.6 ↑2 96.3%
- 3 Claude Opus 4.6 ↑2 96%
- 4
Gemini 3 Pro ↓1 95.8%
- 5 GPT-5.4 ↓3 94.8%
- 6 GPT-5.2 ↑2 86%
- 7
Gemini 3 Flash ↓1 85.6%
- 8
Qwen3-VL-Plus ↓1 85%
- 9 Claude Haiku 4.5 ↑4 81.7%
- 10
Qwen3.5-4B ↑1 76.7%
VQA
23 models we track have a score here.
- 1
Gemini 3.1 Pro 85%
- 2
Qwen3.5-9B ↑7 79.5%
- 3 GPT-5.4 ↓1 78.2%
- 4
Mistral Small 4 ↑11 77.9%
- 5
Qwen3-VL-Plus ↑2 74.7%
- 6
Qwen3.5-4B ↑5 72.4%
- 7
Gemma 4 E4B ↑13 65.3%
- 8 Claude Sonnet 4.6 ↓4 65.2%
- 9 GPT-5 mini ↑3 65%
- 10 Claude Opus 4.6 ↓5 64.4%
6 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official IDP Core Bench leaderboard
Gemini 3.1 Pro on the official leaderboard passed 85% in February 2026
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official IDP Core Bench 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 All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (27 entries).
- Pass Models scored 23 models we track, plus 4 other entries on its leaderboard.
- Note Newest score The newest came 8 months ago, in March 2026, for Gemma 4 E2B. 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 GPT-5.4 (#28).
- Note Headroom Saturated. The best is 89.6%, past 85% since February 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on idp-leaderboard.org: 23 rows for models we track and 4 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.