IDP Leaderboard
Document AI leaderboard combining OCR, table extraction, key information extraction, and visual question answering scores from OlmOCR, OmniDocBench, and IDP Core evaluations.
29rows
overall_scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Overall Score, OlmOCR Overall, OmniDocBench Overall, IDP Core Overall, IDP KIE, IDP OCR, IDP Table, IDP VQA, OlmOCR Table, OlmOCR Math, OlmOCR Old Scans, OmniDocBench Table TEDS, OmniDocBench Formula CDM, OmniDocBench Text Accuracy, OmniDocBench Reading Order Accuracy, Cost per 1K Pages (lower is better)
| Rank | Subject | Overall Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | Nanonets OCR-3 | 85.87 | — | Imported | 2026-05-06 |
| 2 | GPT-5.4 | 83.55 |  GPT-5.4 openai-gpt-5.4 | Imported | 2026-05-06 |
| 3 | Gemini-3-Pro | 82.77 |  Gemini 3 google-gemini-3 | Imported | 2026-05-06 |
| 4 | Gemini-3-Flash | 81.95 | Gemini 3 Flash Preview google-gemini-3-flash-preview | Imported | 2026-05-06 |
| 5 | Nanonets OCR2+ | 81.76 | — | Imported | 2026-05-06 |
| 6 | Gemini 3.1 Pro | 81.58 | Gemini 3.1 Pro Preview google-gemini-3.1-pro-preview | Imported | 2026-05-06 |
| 7 | GPT-5.2 | 81.49 | GPT-5.2 openai-gpt-5.2 | Imported | 2026-05-06 |
| 8 | Claude Sonnet 4.6 | 80.68 |  Claude Sonnet 4.6 anthropic-claude-sonnet-4.6 | Imported | 2026-05-06 |
| 9 | Claude Opus 4.6 | 80.37 | Claude Opus 4.6 anthropic-claude-opus-4.6 | Imported | 2026-05-06 |
| 10 | Qwen3-VL-Plus | 80.06 | — | Imported | 2026-05-06 |
| 11 | Qwen3-VL-235B | 79.57 | — | Imported | 2026-05-06 |
| 12 | Qwen3.5-9B | 76.69 |  Qwen3.5-9B qwen-qwen3.5-9b | Imported | 2026-05-06 |
| 13 | GPT-5-Mini | 75.23 | GPT-5 Mini openai-gpt-5-mini | Imported | 2026-05-06 |
| 14 | Qwen3.5-4B | 72.49 | — | Imported | 2026-05-06 |
| 15 | Mistral Small 4 | 71.50 |  Mistral: Mistral Small 4 mistralai-mistral-small-2603 | Imported | 2026-05-06 |
| 16 | Claude Haiku 4.5 | 71.24 | Claude Haiku 4.5 anthropic-claude-haiku-4.5 | Imported | 2026-05-06 |
| 17 | Ministral-8B | 69.55 | — | Imported | 2026-05-06 |
| 18 | GPT-4.1 | 67.99 | GPT-4.1 openai-gpt-4.1 | Imported | 2026-05-06 |
| 19 | GLM-OCR | 64.18 | — | Imported | 2026-05-06 |
| 20 | Qwen3.5-2B | 62.57 | — | Imported | 2026-05-06 |
| 21 | Qwen3.5-0.8B | 57.78 | — | Imported | 2026-05-06 |
| 22 | GPT-5-Nano | 54.81 | GPT-5 Nano openai-gpt-5-nano | Imported | 2026-05-06 |
| 23 | Gemma-4-E4B-it | 53.90 | — | Imported | 2026-05-06 |
| 24 | Llama-3.2-Vision-11B | 50.78 | — | Imported | 2026-05-06 |
| 25 | Pixtral-12B | 46.52 | — | Imported | 2026-05-06 |
| 26 | Gemma-4-E2B-it | 41.87 | — | Imported | 2026-05-06 |
| 27 | Gemma-3-12B-IT | 0 | Gemma 3 12B google-gemma-3-12b-it | Imported | 2026-05-06 |
| 28 | Datalab Marker | 0 | — | Imported | 2026-05-06 |
| 29 | Qwen-VL-OCR | 0 | — | Imported | 2026-05-06 |
No matching rows.