| { | |
| "leaderboard": [ | |
| { | |
| "model": "FLUX-pro-1.1-Ultra", | |
| "link": "https://bfl.ai/", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2024-11", | |
| "Overall": 75.40, | |
| "Style": 91.36, | |
| "World Knowledge": 91.76, | |
| "Attribute-Overall": 84.97, | |
| "Quantity": 79.26, | |
| "Expression": 68.58, | |
| "Material": 82.98, | |
| "Size": 89.96, | |
| "Shape": 80.59, | |
| "Color": 93.01, | |
| "Action-Overall": 72.43, | |
| "Hand": 67.31, | |
| "Full body": 66.25, | |
| "Animal": 73.19, | |
| "Non Contact": 66.96, | |
| "Contact": 62.07, | |
| "State": 80.53, | |
| "Relationship-Overall": 81.90, | |
| "Composition": 81.89, | |
| "Similarity": 74.04, | |
| "Inclusion": 90.52, | |
| "Comparison": 80.58, | |
| "Compound-Overall": 78.07, | |
| "Imagination": 80.40, | |
| "Feature matching": 72.88, | |
| "Grammar-Overall": 71.94, | |
| "Pronoun Reference": 84.52, | |
| "Consistency": 68.55, | |
| "Negation": 63.73, | |
| "Layout-Overall": 82.62, | |
| "2D": 81.78, | |
| "3D": 83.70, | |
| "Logical Reasoning": 60.92, | |
| "Text": 38.04 | |
| }, | |
| { | |
| "model": "Qwen-Image", | |
| "link": "https://arxiv.org/pdf/2508.02324", | |
| "hf": "https://huggingface.co/Qwen/Qwen-Image", | |
| "open_source": true, | |
| "release_date": "2025-08", | |
| "Overall": 83.94, | |
| "Style": 96.93, | |
| "World Knowledge": 95.09, | |
| "Attribute-Overall": 93.65, | |
| "Quantity": 92.02, | |
| "Expression": 89.86, | |
| "Material": 94.50, | |
| "Size": 89.58, | |
| "Shape": 86.71, | |
| "Color": 97.85, | |
| "Action-Overall": 81.86, | |
| "Hand": 78.53, | |
| "Full body": 81.88, | |
| "Animal": 83.70, | |
| "Non Contact": 83.04, | |
| "Contact": 71.84, | |
| "State": 85.57, | |
| "Relationship-Overall": 83.41, | |
| "Composition": 81.76, | |
| "Similarity": 79.17, | |
| "Inclusion": 88.79, | |
| "Comparison": 85.19, | |
| "Compound-Overall": 81.98, | |
| "Imagination": 82.38, | |
| "Feature matching": 81.07, | |
| "Grammar-Overall": 73.86, | |
| "Pronoun Reference": 90.48, | |
| "Consistency": 78.57, | |
| "Negation": 54.93, | |
| "Layout-Overall": 88.97, | |
| "2D": 91.24, | |
| "3D": 86.05, | |
| "Logical Reasoning": 66.75, | |
| "Text": 76.90 | |
| }, | |
| { | |
| "model": "FLUX-kontext-pro", | |
| "link": "https://bfl.ai/models/flux-kontext", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-05", | |
| "Overall": 78.58, | |
| "Style": 94.83, | |
| "World Knowledge": 93.60, | |
| "Attribute-Overall": 86.24, | |
| "Quantity": 74.47, | |
| "Expression": 75.00, | |
| "Material": 85.47, | |
| "Size": 89.58, | |
| "Shape": 80.63, | |
| "Color": 92.89, | |
| "Action-Overall": 74.44, | |
| "Hand": 73.05, | |
| "Full body": 73.12, | |
| "Animal": 75.00, | |
| "Non Contact": 67.73, | |
| "Contact": 70.40, | |
| "State": 77.98, | |
| "Relationship-Overall": 78.40, | |
| "Composition": 73.85, | |
| "Similarity": 72.08, | |
| "Inclusion": 89.08, | |
| "Comparison": 82.77, | |
| "Compound-Overall": 79.75, | |
| "Imagination": 83.58, | |
| "Feature matching": 71.23, | |
| "Grammar-Overall": 77.05, | |
| "Pronoun Reference": 90.32, | |
| "Consistency": 75.40, | |
| "Negation": 66.90, | |
| "Layout-Overall": 85.46, | |
| "2D": 84.09, | |
| "3D": 87.23, | |
| "Logical Reasoning": 66.26, | |
| "Text": 49.73 | |
| }, | |
| { | |
| "model": "Hunyuan-DiT", | |
| "link": "https://arxiv.org/pdf/2405.08748", | |
| "hf": "https://huggingface.co/Tencent-Hunyuan/HunyuanDiT", | |
| "open_source": true, | |
| "release_date": "2024-05", | |
| "Overall": 54.88, | |
| "Style": 92.94, | |
| "World Knowledge": 80.06, | |
| "Attribute-Overall": 69.47, | |
| "Quantity": 65.43, | |
| "Expression": 52.22, | |
| "Material": 72.14, | |
| "Size": 75.19, | |
| "Shape": 58.22, | |
| "Color": 76.31, | |
| "Action-Overall": 48.80, | |
| "Hand": 39.10, | |
| "Full body": 46.25, | |
| "Animal": 47.46, | |
| "Non Contact": 41.07, | |
| "Contact": 34.48, | |
| "State": 59.58, | |
| "Relationship-Overall": 55.66, | |
| "Composition": 56.89, | |
| "Similarity": 55.45, | |
| "Inclusion": 57.18, | |
| "Comparison": 52.18, | |
| "Compound-Overall": 50.22, | |
| "Imagination": 55.49, | |
| "Feature matching": 38.55, | |
| "Grammar-Overall": 58.76, | |
| "Pronoun Reference": 64.68, | |
| "Consistency": 59.52, | |
| "Negation": 52.82, | |
| "Layout-Overall": 61.43, | |
| "2D": 60.45, | |
| "3D": 62.68, | |
| "Logical Reasoning": 29.85, | |
| "Text": 1.63 | |
| }, | |
| { | |
| "model": "FLUX.1-dev", | |
| "link": "https://bfl.ai/blog/24-08-01-bfl", | |
| "hf": "https://huggingface.co/black-forest-labs/FLUX.1-dev", | |
| "open_source": true, | |
| "release_date": "2024-08", | |
| "Overall": 69.42, | |
| "Style": 89.29, | |
| "World Knowledge": 89.45, | |
| "Attribute-Overall": 79.90, | |
| "Quantity": 73.94, | |
| "Expression": 64.44, | |
| "Material": 80.05, | |
| "Size": 84.47, | |
| "Shape": 71.50, | |
| "Color": 87.47, | |
| "Action-Overall": 64.54, | |
| "Hand": 63.78, | |
| "Full body": 62.50, | |
| "Animal": 65.94, | |
| "Non Contact": 56.70, | |
| "Contact": 56.32, | |
| "State": 69.57, | |
| "Relationship-Overall": 69.40, | |
| "Composition": 65.05, | |
| "Similarity": 66.03, | |
| "Inclusion": 79.60, | |
| "Comparison": 71.60, | |
| "Compound-Overall": 68.46, | |
| "Imagination": 71.10, | |
| "Feature matching": 62.62, | |
| "Grammar-Overall": 70.56, | |
| "Pronoun Reference": 83.33, | |
| "Consistency": 67.46, | |
| "Negation": 61.97, | |
| "Layout-Overall": 77.54, | |
| "2D": 81.21, | |
| "3D": 72.83, | |
| "Logical Reasoning": 54.37, | |
| "Text": 30.71 | |
| }, | |
| { | |
| "model": "FLUX-kontext-max", | |
| "link": "https://bfl.ai/models/flux-kontext", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-05", | |
| "Overall": 80.88, | |
| "Style": 96.51, | |
| "World Knowledge": 93.35, | |
| "Attribute-Overall": 87.45, | |
| "Quantity": 79.79, | |
| "Expression": 76.68, | |
| "Material": 87.35, | |
| "Size": 88.83, | |
| "Shape": 81.51, | |
| "Color": 93.74, | |
| "Action-Overall": 75.52, | |
| "Hand": 73.08, | |
| "Full body": 75.94, | |
| "Animal": 74.28, | |
| "Non Contact": 66.82, | |
| "Contact": 71.55, | |
| "State": 79.76, | |
| "Relationship-Overall": 80.78, | |
| "Composition": 77.30, | |
| "Similarity": 73.05, | |
| "Inclusion": 89.94, | |
| "Comparison": 85.44, | |
| "Compound-Overall": 82.24, | |
| "Imagination": 84.75, | |
| "Feature matching": 76.65, | |
| "Grammar-Overall": 79.34, | |
| "Pronoun Reference": 90.08, | |
| "Consistency": 76.61, | |
| "Negation": 72.18, | |
| "Layout-Overall": 87.58, | |
| "2D": 85.73, | |
| "3D": 89.96, | |
| "Logical Reasoning": 71.12, | |
| "Text": 54.89 | |
| }, | |
| { | |
| "model": "Recraft", | |
| "link": "https://www.recraft.ai/docs#generate-image", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2024-12", | |
| "Overall": 60.93, | |
| "Style": 87.13, | |
| "World Knowledge": 86.99, | |
| "Attribute-Overall": 73.23, | |
| "Quantity": 56.38, | |
| "Expression": 57.22, | |
| "Material": 72.82, | |
| "Size": 76.89, | |
| "Shape": 63.64, | |
| "Color": 83.07, | |
| "Action-Overall": 51.77, | |
| "Hand": 40.06, | |
| "Full body": 54.37, | |
| "Animal": 55.07, | |
| "Non Contact": 45.09, | |
| "Contact": 37.36, | |
| "State": 60.08, | |
| "Relationship-Overall": 55.82, | |
| "Composition": 51.79, | |
| "Similarity": 46.47, | |
| "Inclusion": 66.09, | |
| "Comparison": 61.89, | |
| "Compound-Overall": 49.56, | |
| "Imagination": 50.21, | |
| "Feature matching": 48.13, | |
| "Grammar-Overall": 60.28, | |
| "Pronoun Reference": 73.41, | |
| "Consistency": 55.56, | |
| "Negation": 52.82, | |
| "Layout-Overall": 63.81, | |
| "2D": 65.96, | |
| "3D": 61.05, | |
| "Logical Reasoning": 34.22, | |
| "Text": 46.47 | |
| }, | |
| { | |
| "model": "wan2.2-t2i-plus", | |
| "link": "https://help.aliyun.com/zh/model-studio/text-to-image-v2-api-reference", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-07", | |
| "Overall": 68.76, | |
| "Style": 90.28, | |
| "World Knowledge": 87.57, | |
| "Attribute-Overall": 81.08, | |
| "Quantity": 78.19, | |
| "Expression": 69.17, | |
| "Material": 80.42, | |
| "Size": 82.77, | |
| "Shape": 73.60, | |
| "Color": 88.10, | |
| "Action-Overall": 66.49, | |
| "Hand": 64.10, | |
| "Full body": 60.94, | |
| "Animal": 70.29, | |
| "Non Contact": 59.38, | |
| "Contact": 55.46, | |
| "State": 73.32, | |
| "Relationship-Overall": 72.79, | |
| "Composition": 69.13, | |
| "Similarity": 66.67, | |
| "Inclusion": 81.03, | |
| "Comparison": 77.43, | |
| "Compound-Overall": 71.73, | |
| "Imagination": 74.16, | |
| "Feature matching": 66.36, | |
| "Grammar-Overall": 70.18, | |
| "Pronoun Reference": 86.90, | |
| "Consistency": 61.11, | |
| "Negation": 63.38, | |
| "Layout-Overall": 79.13, | |
| "2D": 82.34, | |
| "3D": 75.00, | |
| "Logical Reasoning": 55.58, | |
| "Text": 12.77 | |
| }, | |
| { | |
| "model": "Nano Banana", | |
| "link": "https://ainanobanana.io/", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-08", | |
| "Overall": 88.82, | |
| "Style": 98.83, | |
| "World Knowledge": 95.78, | |
| "Attribute-Overall": 93.06, | |
| "Quantity": 88.24, | |
| "Expression": 86.09, | |
| "Material": 93.05, | |
| "Size": 93.70, | |
| "Shape": 88.73, | |
| "Color": 97.31, | |
| "Action-Overall": 83.93, | |
| "Hand": 84.57, | |
| "Full body": 84.95, | |
| "Animal": 81.16, | |
| "Non Contact": 83.41, | |
| "Contact": 78.16, | |
| "State": 86.28, | |
| "Relationship-Overall": 91.59, | |
| "Composition": 90.98, | |
| "Similarity": 91.32, | |
| "Inclusion": 92.80, | |
| "Comparison": 91.91, | |
| "Compound-Overall": 90.63, | |
| "Imagination": 92.15, | |
| "Feature matching": 87.23, | |
| "Grammar-Overall": 89.33, | |
| "Pronoun Reference": 94.84, | |
| "Consistency": 89.24, | |
| "Negation": 84.51, | |
| "Layout-Overall": 94.04, | |
| "2D": 94.77, | |
| "3D": 93.12, | |
| "Logical Reasoning": 81.27, | |
| "Text": 69.75 | |
| }, | |
| { | |
| "model": "GPT-4o", | |
| "link": "https://platform.openai.com/docs/guides/image-generation", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-03", | |
| "Overall": 92.63, | |
| "Style": 99.08, | |
| "World Knowledge": 97.95, | |
| "Attribute-Overall": 93.53, | |
| "Quantity": 86.70, | |
| "Expression": 93.44, | |
| "Material": 92.45, | |
| "Size": 94.89, | |
| "Shape": 92.48, | |
| "Color": 94.95, | |
| "Action-Overall": 87.78, | |
| "Hand": 89.94, | |
| "Full body": 87.19, | |
| "Animal": 90.94, | |
| "Non Contact": 89.29, | |
| "Contact": 83.05, | |
| "State": 87.75, | |
| "Relationship-Overall": 91.13, | |
| "Composition": 89.18, | |
| "Similarity": 90.71, | |
| "Inclusion": 96.84, | |
| "Comparison": 90.29, | |
| "Compound-Overall": 93.99, | |
| "Imagination": 94.39, | |
| "Feature matching": 93.10, | |
| "Grammar-Overall": 94.46, | |
| "Pronoun Reference": 95.97, | |
| "Consistency": 91.67, | |
| "Negation": 95.65, | |
| "Layout-Overall": 93.59, | |
| "2D": 94.29, | |
| "3D": 92.70, | |
| "Logical Reasoning": 91.02, | |
| "Text": 83.79 | |
| }, | |
| { | |
| "model": "Seedream-3.0", | |
| "link": "https://www.byteplus.com/en/product/Seedream", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-06", | |
| "Overall": 80.99, | |
| "Style": 97.18, | |
| "World Knowledge": 93.79, | |
| "Attribute-Overall": 91.90, | |
| "Quantity": 83.51, | |
| "Expression": 81.25, | |
| "Material": 93.07, | |
| "Size": 88.26, | |
| "Shape": 90.03, | |
| "Color": 97.48, | |
| "Action-Overall": 79.94, | |
| "Hand": 77.88, | |
| "Full body": 84.69, | |
| "Animal": 78.26, | |
| "Non Contact": 74.11, | |
| "Contact": 71.84, | |
| "State": 83.60, | |
| "Relationship-Overall": 83.41, | |
| "Composition": 81.63, | |
| "Similarity": 79.17, | |
| "Inclusion": 87.64, | |
| "Comparison": 86.41, | |
| "Compound-Overall": 81.03, | |
| "Imagination": 80.49, | |
| "Feature matching": 82.24, | |
| "Grammar-Overall": 75.13, | |
| "Pronoun Reference": 90.48, | |
| "Consistency": 80.56, | |
| "Negation": 56.69, | |
| "Layout-Overall": 88.41, | |
| "2D": 87.85, | |
| "3D": 89.13, | |
| "Logical Reasoning": 62.62, | |
| "Text": 56.52 | |
| }, | |
| { | |
| "model": "Imagen-4.0-Fast-preview-06-06", | |
| "link": "https://deepmind.google/models/imagen/", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-06", | |
| "Overall": 81.54, | |
| "Style": 93.77, | |
| "World Knowledge": 93.64, | |
| "Attribute-Overall": 90.33, | |
| "Quantity": 78.72, | |
| "Expression": 78.89, | |
| "Material": 91.11, | |
| "Size": 90.15, | |
| "Shape": 86.89, | |
| "Color": 96.33, | |
| "Action-Overall": 80.18, | |
| "Hand": 82.05, | |
| "Full body": 84.06, | |
| "Animal": 81.88, | |
| "Non Contact": 75.00, | |
| "Contact": 74.71, | |
| "State": 80.93, | |
| "Relationship-Overall": 84.05, | |
| "Composition": 82.53, | |
| "Similarity": 80.13, | |
| "Inclusion": 92.82, | |
| "Comparison": 82.52, | |
| "Compound-Overall": 84.01, | |
| "Imagination": 86.18, | |
| "Feature matching": 79.21, | |
| "Grammar-Overall": 79.57, | |
| "Pronoun Reference": 91.27, | |
| "Consistency": 81.35, | |
| "Negation": 67.61, | |
| "Layout-Overall": 90.48, | |
| "2D": 90.11, | |
| "3D": 90.94, | |
| "Logical Reasoning": 67.72, | |
| "Text": 51.63 | |
| }, | |
| { | |
| "model": "Imagen-4.0-Ultra-preview-06-06", | |
| "link": "https://deepmind.google/models/imagen/", | |
| "hf": "-", | |
| "open_source": false, | |
| "release_date": "2025-06", | |
| "Overall": 90.95, | |
| "Style": 97.67, | |
| "World Knowledge": 98.26, | |
| "Attribute-Overall": 93.21, | |
| "Quantity": 89.84, | |
| "Expression": 83.17, | |
| "Material": 94.20, | |
| "Size": 94.69, | |
| "Shape": 89.86, | |
| "Color": 97.22, | |
| "Action-Overall": 86.91, | |
| "Hand": 89.10, | |
| "Full body": 86.56, | |
| "Animal": 85.14, | |
| "Non Contact": 86.61, | |
| "Contact": 81.84, | |
| "State": 88.63, | |
| "Relationship-Overall": 90.57, | |
| "Composition": 90.05, | |
| "Similarity": 84.62, | |
| "Inclusion": 94.52, | |
| "Comparison": 92.72, | |
| "Compound-Overall": 91.42, | |
| "Imagination": 92.82, | |
| "Feature matching": 88.32, | |
| "Grammar-Overall": 88.07, | |
| "Pronoun Reference": 96.83, | |
| "Consistency": 87.70, | |
| "Negation": 80.63, | |
| "Layout-Overall": 93.49, | |
| "2D": 92.64, | |
| "3D": 94.57, | |
| "Logical Reasoning": 83.50, | |
| "Text": 86.41 | |
| } | |
| ] | |
| } | |