| ParseBench: A Document Parsing Benchmark for AI Agents | ✓ Link | 90.2 | 94.18 | 93.37 | 84.09 | 87.12 | 92.25 | LlamaParse Agentic Plus | 2026-04-09 |
| ParseBench: A Document Parsing Benchmark for AI Agents | ✓ Link | 87.01 | 88.68 | 88.88 | 84.25 | 81.44 | 91.78 | LlamaParse Agentic | 2026-04-09 |
| ParseBench: A Document Parsing Benchmark for AI Agents | ✓ Link | 80.61 | 77.91 | 84.19 | 83.77 | 67.29 | 89.87 | LlamaParse Cost Effective | 2026-04-09 |
| ParseBench Leaderboard | ✓ Link | 79.85 | 70.89 | 94.25 | 65.33 | 77.04 | 91.72 | Anthropic Opus 5.5 (Effort High) | 2026-09-22 |
| ParseBench Leaderboard | | 78.92 | 67.06 | 91.52 | 68.3 | 76.52 | 91.19 | Anthropic Fable 5.1 | 2026-09-01 |
| ParseBench Leaderboard | ✓ Link | 78.01 | 64.12 | 93.86 | 63.2 | 77.04 | 91.81 | Anthropic Opus 5.5 | 2026-09-22 |
| ParseBench Leaderboard | | 77.23 | 64.89 | 86.44 | 74.28 | 71.68 | 88.84 | rakedoc-nano | 2026-08-29 |
| ParseBench Leaderboard | | 76.77 | 68.78 | 85.89 | 75.21 | 66.05 | 87.9 | aeddix-alpine-ocr (KDL pipeline) | 2026-10-06 |
| ParseBench Leaderboard | | 76.69 | 65.19 | 86.1 | 74.14 | 70.64 | 87.37 | florin-parser-nano | 2026-08-17 |
| ParseBench Leaderboard | ✓ Link | 75.35 | 68.68 | 91.39 | 51.35 | 77.16 | 88.16 | OpenAI GPT-5.6 Sol (Reasoning High) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 75.1 | 66.1 | 83.8 | 68.3 | 67.6 | 89.9 | LightOnOCR-3-4B | 2026-10-08 |
| ParseBench Leaderboard | | 75.05 | 64.79 | 91.5 | 59.77 | 68.31 | 90.87 | Google Gemini 3 Flash (Thinking High) | 2025-12-17 |
| ParseBench Leaderboard | | 75.0 | 61.56 | 91.01 | 65.79 | 67.95 | 88.67 | Google Gemini 3 Flash (Thinking Medium) | 2025-12-17 |
| ParseBench Leaderboard | ✓ Link | 74.6 | 64.7 | 84.5 | 68.2 | 66.7 | 88.9 | LightOnOCR-3-0.8B | 2026-10-08 |
| ParseBench Leaderboard | ✓ Link | 74.47 | 53.52 | 92.69 | 59.27 | 75.21 | 91.68 | Anthropic Opus 5.5 (Effort Low) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 74.28 | 61.3 | 86.4 | 74.9 | 59.1 | 89.7 | Infinity-Parser2-Pro | 2026-07-08 |
| ParseBench Leaderboard | | 74.02 | 45.27 | 90.44 | 67.6 | 77.85 | 88.92 | Google Gemini 3.5 Flash (Thinking High) | 2026-05-19 |
| ParseBench Leaderboard | ✓ Link | 73.25 | 55.56 | 82.88 | 80.61 | 57.7 | 89.52 | Infinity-Parser2-Flash | 2026-07-08 |
| ParseBench Leaderboard | ✓ Link | 72.78 | 61.64 | 77.59 | 79.3 | 57.49 | 87.88 | MinerU2.5-Pro-2605-1.2B | 2026-04-06 |
| ParseBench Leaderboard | | 72.38 | 58.61 | 85.73 | 74.75 | 55.06 | 87.75 | KDL-Frontier-Parser-nano | 2026-06-10 |
| ParseBench Leaderboard | ✓ Link | 72.08 | 32.34 | 89.12 | 72.91 | 76.36 | 89.68 | Google Gemini 3.8 Flash (Thinking High) | 2026-09-02 |
| ParseBench Leaderboard | ✓ Link | 71.4 | 57.3 | 84.8 | 61.7 | 64.5 | 88.6 | LightOnOCR-3-1B | 2026-10-08 |
| ParseBench Leaderboard | ✓ Link | 71.27 | 32.07 | 90.16 | 70.56 | 74.98 | 88.6 | Google Gemini 3.7 Flash (Thinking High) | 2026-08-13 |
| ParseBench Leaderboard | | 71.04 | 64.83 | 89.85 | 55.97 | 58.35 | 86.19 | Google Gemini 3 Flash (Thinking Minimal) | 2025-12-17 |
| ParseBench Leaderboard | ✓ Link | 70.79 | 69.17 | 66.82 | 69.9 | 59.77 | 88.28 | Qwen 3.8 27B (Thinking) | 2026-08-05 |
| ParseBench Leaderboard | ✓ Link | 70.78 | 52.21 | 89.79 | 49.24 | 72.62 | 90.02 | Anthropic Fable 5 | 2026-06-09 |
| ParseBench Leaderboard | ✓ Link | 70.78 | 35.11 | 88.66 | 68.63 | 73.05 | 88.45 | Google Gemini 3.7 Flash (Thinking Medium) | 2026-08-13 |
| ParseBench Leaderboard | ✓ Link | 70.75 | 54.96 | 89.18 | 41.78 | 79.49 | 88.34 | GLM-5.3 Flash | 2026-08-26 |
| ParseBench Leaderboard | ✓ Link | 70.7 | 32.72 | 89.08 | 71.5 | 71.8 | 88.42 | Google Gemini 3.8 Flash (Thinking Medium) | 2026-09-02 |
| ParseBench Leaderboard | ✓ Link | 70.67 | 35.08 | 93.17 | 59.99 | 75.52 | 89.6 | OpenAI GPT-6 Astra (Reasoning Low) | 2026-09-03 |
| ParseBench Leaderboard | | 70.48 | 37.16 | 92.44 | 58.76 | 74.78 | 89.25 | OpenAI GPT-6.1 Sol (Reasoning Low) | 2026-09-29 |
| ParseBench Leaderboard | ✓ Link | 70.18 | 37.69 | 91.04 | 58.92 | 71.63 | 91.64 | Anthropic Sonnet 5.5 | 2026-09-28 |
| ParseBench Leaderboard | ✓ Link | 70.15 | 35.09 | 88.18 | 72.24 | 66.99 | 88.25 | Google Gemini 3.8 Flash (Thinking Low) | 2026-09-02 |
| ParseBench Leaderboard | ✓ Link | 70.1 | 65.1 | 89.2 | 51.2 | 61.4 | 83.7 | Chandra-ocr-2 | 2026-03-16 |
| ParseBench Leaderboard | | 70.03 | 34.53 | 89.3 | 64.98 | 73.11 | 88.21 | Google Gemini 3.6 Flash (Thinking High) | 2026-07-21 |
| ParseBench Leaderboard | | 69.92 | 44.01 | 91.12 | 65.14 | 59.14 | 90.19 | Google Gemini 3.5 Flash (Thinking Medium) | 2026-05-19 |
| ParseBench Leaderboard | | 69.69 | 48.05 | 92.63 | 52.06 | 67.91 | 87.81 | OpenAI GPT-6 Sol (Reasoning High) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 69.57 | 31.84 | 87.96 | 70.71 | 68.89 | 88.44 | Google Gemini 3.7 Flash (Thinking Low) | 2026-08-13 |
| ParseBench Leaderboard | | 69.14 | 41.13 | 91.0 | 70.99 | 52.43 | 90.16 | Google Gemini 3.1 Pro | 2026-02-19 |
| ParseBench Leaderboard | ✓ Link | 68.39 | 67.86 | 86.72 | 31.82 | 68.47 | 87.06 | OpenAI GPT-5.6 Terra (Reasoning Medium) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 68.34 | 49.16 | 88.27 | 52.73 | 64.95 | 86.57 | OpenAI GPT-5.6 Luna (Reasoning Max) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 68.24 | 49.44 | 89.42 | 48.69 | 66.51 | 87.13 | OpenAI GPT-5.6 Luna (Reasoning XHigh) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 68.23 | 40.11 | 73.94 | 71.17 | 66.37 | 89.55 | Mistral OCR 4 (Annotation) | 2026-06-23 |
| ParseBench Leaderboard | | 68.19 | 46.46 | 90.27 | 49.05 | 67.88 | 87.3 | OpenAI GPT-6 Sol (Reasoning Medium) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 68.07 | 67.71 | 86.91 | 30.97 | 67.92 | 86.83 | OpenAI GPT-5.6 Terra (Reasoning Low) | 2026-07-09 |
| ParseBench Leaderboard | | 67.76 | 65.53 | 90.05 | 36.28 | 60.12 | 86.81 | OpenAI GPT-5.5 (Reasoning Medium) | 2026-04-23 |
| ParseBench Leaderboard | ✓ Link | 67.43 | 54.24 | 67.77 | 77.8 | 54.64 | 82.71 | PaddleOCR-VL-1.6 | 2026-06-02 |
| ParseBench Leaderboard | ✓ Link | 67.37 | 49.25 | 88.75 | 44.74 | 67.14 | 86.99 | OpenAI GPT-5.6 Luna (Reasoning High) | 2026-07-09 |
| ParseBench Leaderboard | | 66.78 | 31.0 | 89.47 | 67.27 | 58.46 | 87.7 | Google Gemini 3.6 Flash (Thinking Medium) | 2026-07-21 |
| ParseBench Leaderboard | | 66.1 | 42.02 | 89.49 | 45.27 | 67.32 | 86.38 | OpenAI GPT-6 Sol (Reasoning None) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 65.95 | 47.62 | 67.38 | 77.78 | 54.27 | 82.72 | PaddleOCR-VL-1.5 | 2026-01-29 |
| ParseBench Leaderboard | | 65.77 | 45.71 | 87.42 | 49.48 | 61.81 | 84.44 | OpenAI GPT-6 Luna (Reasoning Max) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 65.72 | 51.11 | 82.82 | 33.87 | 73.48 | 87.33 | DeepSeek V4.1 Flash (Thinking High) | 2026-09-17 |
| ParseBench Leaderboard | | 65.6 | 24.76 | 85.5 | 69.94 | 64.49 | 83.3 | Google Gemini 3.6 Flash (Thinking Minimal) | 2026-07-21 |
| ParseBench Leaderboard | ✓ Link | 64.83 | 21.95 | 82.68 | 71.59 | 61.36 | 86.57 | Surya OCR 2 | 2026-05-27 |
| ParseBench Leaderboard | | 64.39 | 59.11 | 89.31 | 21.9 | 64.46 | 87.17 | OpenAI GPT-5.5 (Reasoning None) | 2026-04-23 |
| ParseBench Leaderboard | | 64.32 | 42.38 | 88.01 | 40.86 | 65.34 | 85.0 | OpenAI GPT-6 Luna (Reasoning XHigh) | 2026-09-22 |
| ParseBench Leaderboard | | 64.21 | 44.28 | 87.97 | 39.97 | 63.58 | 85.27 | OpenAI GPT-6 Luna (Reasoning High) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 64.17 | 63.75 | 86.23 | 28.35 | 59.96 | 82.56 | OpenAI GPT-5.6 Terra (Reasoning None) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 63.86 | 55.98 | 65.45 | 63.72 | 47.73 | 86.44 | HunyuanOCR 1.5 | 2026-07-06 |
| ParseBench Leaderboard | | 63.7 | 49.75 | 89.65 | 18.69 | 71.38 | 89.02 | Anthropic Opus 4.8 | 2026-05-28 |
| ParseBench Leaderboard | ✓ Link | 63.58 | 46.91 | 85.32 | 34.6 | 65.8 | 85.27 | OpenAI GPT-5.6 Luna (Reasoning Medium) | 2026-07-09 |
| ParseBench Leaderboard | | 63.34 | 55.84 | 87.17 | 13.99 | 69.42 | 90.26 | Anthropic Opus 4.7 | 2026-04-16 |
| ParseBench Leaderboard | | 63.09 | 18.74 | 86.24 | 54.18 | 68.09 | 88.22 | Google Gemini 3.5 Flash (Thinking Minimal) | 2026-05-19 |
| ParseBench Leaderboard | ✓ Link | 62.4 | 15.0 | 80.6 | 57.4 | 69.3 | 89.9 | Gemma-4-31B-it | 2026-07-02 |
| ParseBench Leaderboard | | 62.38 | 42.97 | 86.74 | 36.14 | 61.13 | 84.94 | OpenAI GPT-6 Luna (Reasoning Medium) | 2026-09-22 |
| ParseBench Leaderboard | | 62.23 | 65.22 | 83.89 | 16.95 | 59.52 | 85.57 | OpenAI GPT-5.4 (Reasoning None) | 2026-03-05 |
| ParseBench Leaderboard | | 62.13 | 60.53 | 86.65 | 12.03 | 64.93 | 86.51 | Anthropic Sonnet 5 | 2026-06-30 |
| ParseBench Leaderboard | ✓ Link | 62.12 | 59.57 | 89.34 | 25.09 | 50.42 | 86.17 | OpenAI GPT-5.6 Sol (Reasoning None) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 61.97 | 28.18 | 74.61 | 55.18 | 64.23 | 87.63 | Qwen3-VL-8B-Instruct | 2025-11-26 |
| ParseBench Leaderboard | ✓ Link | 61.21 | 16.59 | 84.52 | 63.06 | 55.29 | 86.57 | Qwen 3.8 27B | 2026-08-05 |
| ParseBench Leaderboard | | 60.82 | 11.43 | 90.29 | 69.7 | 45.56 | 87.11 | Google Gemini 3.1 Pro (Thinking Low) | 2026-02-19 |
| ParseBench Leaderboard | ✓ Link | 60.68 | 2.35 | 73.94 | 71.17 | 66.37 | 89.55 | Mistral OCR 4 | 2026-06-23 |
| ParseBench Leaderboard | | 60.64 | 19.27 | 83.31 | 59.75 | 54.03 | 86.85 | Google Gemini 3.5 Flash Lite (Thinking High) | 2026-07-21 |
| ParseBench Leaderboard | ✓ Link | 60.18 | 38.95 | 81.67 | 30.26 | 64.75 | 85.28 | OpenAI GPT-5.6 Luna (Reasoning Low) | 2026-07-09 |
| ParseBench Leaderboard | | 59.32 | 39.05 | 84.2 | 28.51 | 59.9 | 84.95 | OpenAI GPT-6 Luna (Reasoning Low) | 2026-09-22 |
| ParseBench Leaderboard | ✓ Link | 58.67 | 9.97 | 72.74 | 80.14 | 45.13 | 85.39 | TeleOCR | 2026-08-13 |
| ParseBench Leaderboard | ✓ Link | 58.5 | 14.2 | 70.0 | 59.2 | 65.1 | 83.8 | Gemma-4-26B-A4B-it | 2026-07-02 |
| ParseBench Leaderboard | | 58.32 | 9.92 | 85.48 | 48.38 | 58.38 | 89.46 | Google Gemini 3.1 Flash Lite | 2026-03-03 |
| ParseBench Leaderboard | | 58.13 | 14.35 | 83.43 | 49.05 | 56.17 | 87.67 | Google Gemini 3.1 Flash Lite (Thinking Medium) | 2026-03-03 |
| ParseBench Leaderboard | | 57.4 | 10.13 | 86.85 | 34.14 | 65.64 | 90.24 | Anthropic Haiku 5.5 | 2026-10-07 |
| ParseBench Leaderboard | | 57.29 | 30.89 | 76.76 | 38.93 | 56.61 | 83.28 | Google Gemini 3.1 Flash Lite (Thinking High) | 2026-03-03 |
| ParseBench Leaderboard | | 57.15 | 6.26 | 73.48 | 66.03 | 57.74 | 82.22 | Google Gemini 3.5 Flash Lite | 2026-07-21 |
| ParseBench Leaderboard | ✓ Link | 56.57 | 20.46 | 79.65 | 28.61 | 66.05 | 88.09 | DeepSeek V4.1 Flash (No Thinking) | 2026-09-17 |
| ParseBench Leaderboard | ✓ Link | 56.32 | 28.32 | 81.3 | 25.43 | 63.9 | 82.65 | OpenAI GPT-5.6 Luna (Reasoning None) | 2026-07-09 |
| ParseBench Leaderboard | ✓ Link | 55.79 | 0.95 | 85.15 | 55.81 | 46.99 | 90.03 | Dots.mocr | 2026-03-13 |
| ParseBench Leaderboard | | 54.3 | 6.23 | 72.85 | 65.55 | 48.05 | 78.8 | Google Gemini 3.5 Flash Lite (Thinking Medium) | 2026-07-21 |
| ParseBench Leaderboard | | 54.07 | 13.49 | 86.52 | 16.47 | 64.19 | 89.7 | Anthropic Opus 4.6 | 2026-02-05 |
| ParseBench Leaderboard | ✓ Link | 54.04 | 0.79 | 73.77 | 67.76 | 45.67 | 82.21 | HPD-Parsing | 2026-07-21 |
| ParseBench Leaderboard | | 53.12 | 27.39 | 78.7 | 12.3 | 62.36 | 84.83 | Anthropic Haiku 4.5 (Thinking) | 2025-10-15 |
| ParseBench Leaderboard | ✓ Link | 53.08 | 0.82 | 74.7 | 63.37 | 47.91 | 78.59 | Falcon-OCR | 2026-03-28 |
| ParseBench Leaderboard | | 52.93 | 15.48 | 81.42 | 22.85 | 59.84 | 85.07 | OpenAI GPT-6 Luna (Reasoning None) | 2026-09-22 |
| ParseBench Leaderboard | | 51.52 | 38.96 | 74.6 | 13.03 | 45.35 | 85.68 | OpenAI GPT-5 Mini (Reasoning Medium) | 2025-08-07 |
| ParseBench Leaderboard | ✓ Link | 50.65 | 52.76 | 66.41 | 66.11 | 1.03 | 66.93 | Docling-models | 2024-08-19 |
| ParseBench Leaderboard | ✓ Link | 48.51 | 31.47 | 70.44 | 0.0 | 59.69 | 80.97 | Nemotron-3-Nano-Omni-30B-A3B (Thinking) | 2026-04-27 |
| ParseBench Leaderboard | | 48.0 | 13.5 | 75.5 | 0.0 | 63.2 | 87.8 | LightOnOCR-2-1B | 2026-01-20 |
| ParseBench Leaderboard | | 46.83 | 30.13 | 69.82 | 6.15 | 45.77 | 82.3 | OpenAI GPT-5 Mini (Reasoning Minimal) | 2025-08-07 |
| ParseBench Leaderboard | ✓ Link | 46.2 | 0.9 | 72.3 | 20.6 | 53.8 | 83.5 | Qianfan-OCR | 2026-03-11 |
| ParseBench Leaderboard | ✓ Link | 46.17 | 1.34 | 70.21 | 71.52 | 0.97 | 86.81 | Unlimited-OCR | 2026-06-22 |
| ParseBench Leaderboard | | 45.93 | 7.52 | 74.9 | 0.0 | 60.64 | 86.58 | Jina-OCR-v1 | 2026-09-02 |
| ParseBench Leaderboard | ✓ Link | 45.9 | 1.1 | 71.7 | 71.5 | 4.5 | 80.8 | MinerU2.5-2509-1.2B | 2025-09-26 |
| ParseBench Leaderboard | | 45.17 | 13.77 | 77.21 | 6.72 | 49.39 | 78.74 | Anthropic Haiku 4.5 (Disable Thinking) | 2025-10-15 |
| ParseBench Leaderboard | ✓ Link | 44.1 | 5.1 | 19.1 | 47.4 | 58.3 | 90.7 | Qwen3.6-35B-A3B | 2026-04-21 |
| ParseBench Leaderboard | | 43.35 | 17.57 | 60.16 | 6.41 | 54.56 | 78.05 | OpenAI GPT-5.4 Nano | 2026-03-17 |
| ParseBench Leaderboard | ✓ Link | 42.63 | 1.19 | 66.51 | 74.54 | 1.46 | 69.46 | MinerU-Diffusion-V1-0320-2.5B | 2026-03-23 |
| ParseBench Leaderboard | ✓ Link | 41.2 | 1.1 | 61.7 | 7.0 | 54.0 | 82.0 | DeepSeek-OCR-2 | 2026-01-28 |
| ParseBench Leaderboard | ✓ Link | 40.9 | 0.9 | 67.1 | 0.0 | 54.0 | 82.7 | PaddleOCR-VL | 2025-10-16 |
| ParseBench Leaderboard | | 40.84 | 1.01 | 77.37 | 0.0 | 37.35 | 88.47 | OvisOCR2 | 2026-07-15 |
| ParseBench Leaderboard | ✓ Link | 40.5 | 7.7 | 25.0 | 35.8 | 52.5 | 81.4 | Gemma-4-E4B-it | 2026-07-02 |
| ParseBench Leaderboard | | 40.43 | 6.7 | 68.38 | 4.78 | 45.26 | 77.04 | Amazon Nova 2 Lite | 2025-12-02 |
| ParseBench Leaderboard | | 39.45 | 47.47 | 63.81 | 0.0 | 21.52 | 64.43 | Granite Vision 4.1 4B | 2026-04-16 |
| ParseBench Leaderboard | ✓ Link | 38.93 | 0.86 | 70.41 | 0.0 | 46.28 | 77.11 | WeVisDoc-2B | 2026-09-17 |
| ParseBench Leaderboard | ✓ Link | 36.15 | 0.79 | 65.86 | 0.0 | 43.65 | 70.45 | WeVisDoc-4B | 2026-09-17 |
| ParseBench Leaderboard | ✓ Link | 35.4 | 2.5 | 8.0 | 19.7 | 57.8 | 88.9 | Qwen3.5-4B | 2026-02-16 |
| ParseBench Leaderboard | ✓ Link | 31.9 | 5.4 | 9.9 | 31.8 | 22.0 | 90.5 | Qwen3.5-9B | 2026-02-16 |
| ParseBench Leaderboard | ✓ Link | 29.6 | 1.7 | 66.1 | 0.0 | 2.3 | 78.0 | GLM-OCR | 2026-03-11 |
| ParseBench Leaderboard | ✓ Link | 28.4 | 0.4 | 1.5 | 15.0 | 43.1 | 82.0 | Qwen3.5-0.8B | 2026-02-16 |
| ParseBench Leaderboard | ✓ Link | 27.3 | 0.1 | 0.0 | 18.3 | 31.1 | 87.2 | Qwen3.5-2B | 2026-02-16 |