Hidream_t2i_human_preference
收藏魔搭社区2025-12-04 更新2025-05-17 收录
下载链接：
https://modelscope.cn/datasets/AI-ModelScope/Hidream_t2i_human_preference
下载链接
链接失效反馈官方服务：
资源简介：
<style>
.vertical-container {
display: flex;
flex-direction: column;
gap: 60px;
}
.image-container img {
max-height: 250px; /* Set the desired height */
margin:0;
object-fit: contain; /* Ensures the aspect ratio is maintained */
width: auto; /* Adjust width automatically based on height */
box-sizing: content-box;
}
.image-container {
display: flex; /* Aligns images side by side */
justify-content: space-around; /* Space them evenly */
align-items: center; /* Align them vertically */
gap: .5rem
}
.container {
width: 90%;
margin: 0 auto;
}
.text-center {
text-align: center;
}
.score-amount {
margin: 0;
margin-top: 10px;
}
.score-percentage {Score:
font-size: 12px;
font-weight: semi-bold;
}
</style>
# Rapidata Hidream I1 full Preference
<a href="https://www.rapidata.ai">
<img src="https://cdn-uploads.huggingface.co/production/uploads/66f5624c42b853e73e0738eb/jfxR79bOztqaC6_yNNnGU.jpeg" width="400" alt="Dataset visualization">
</a>
This T2I dataset contains over 195k human responses from over 38k individual annotators, collected in just ~1 Day using the [Rapidata Python API](https://docs.rapidata.ai), accessible to anyone and ideal for large scale evaluation.
Evaluating Hidream I1 full across three categories: preference, coherence, and alignment.
Explore our latest model rankings on our [website](https://www.rapidata.ai/benchmark).
If you get value from this dataset and would like to see more in the future, please consider liking it ❤️
## Overview
This T2I dataset contains over 195k human responses from over 38k individual annotators, collected in just ~1 Day.
Evaluating Hidream I1 full across three categories: preference, coherence, and alignment.
The evaluation consists of 1v1 comparisons between Hidream I1 full and 14 other models: Halfmoon-4-4-2025, OpenAI 4o-26-3-25, Ideogram V2, Recraft V2, Lumina-15-2-25, Frames-23-1-25, Imagen-3, Flux-1.1-pro, Flux-1-pro, DALL-E 3, Midjourney-5.2, Stable Diffusion 3, Aurora and Janus-7b.
> **Note:** The number following the model name (e.g., Halfmoon-4-4-2025) represents the date (April 4, 2025) on which the images were generated to give an understanding of what model version was used.
## Alignment
The alignment score quantifies how well an video matches its prompt. Users were asked: "Which image matches the description better?".
<div class="vertical-container">
<div class="container">
<div class="text-center">
<q>A person under a car.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full </h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/w3pNtIgrKMMULW77oBqJ4.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux-1.1-pro </h3>
<div class="score-percentage">Score: 0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/0g6ipnW8J4PX0VXnIWK2k.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>A chair on a cat.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full</h3>
<div class="score-percentage">Score: 0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/OA9nYfpfewJG98iJ-3G2Y.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">OpenAI 4o-26-3-25</h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/MMOdg5L4cJW7wSLxT-QO0.jpeg" width=500>
</div>
</div>
</div>
</div>
## Coherence
The coherence score measures whether the generated video is logically consistent and free from artifacts or visual glitches. Without seeing the original prompt, users were asked: "Which image has **more** glitches and is **more** likely to be AI generated?"
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full </h3>
<div class="score-percentage">Glitch Rating: 5.84%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/GSqTj1Ole9l9UaM1cnNLc.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Janus-7B </h3>
<div class="score-percentage">Glitch Rating: 94.16%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/v2b-vDkUG1Y6paClTOFuK.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full </h3>
<div class="score-percentage">Glitch Rating: 92.46%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/y5DcMgqhxSqH3mbKaIDJP.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Imagen-3</h3>
<div class="score-percentage">Glitch Rating: 7.54%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/REveLQYhVXC51gWEM7YU6.jpeg" width=500>
</div>
</div>
</div>
</div>
## Preference
The preference score reflects how visually appealing participants found each image, independent of the prompt. Users were asked: "Which image do you prefer?"
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full</h3>
<div class="score-percentage">Score: 70.04%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/hH1ewWQ4tYtqaiSd7qZF7.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Stable Diffusion 3</h3>
<div class="score-percentage">Score: 29.96%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/Uu5pl6zBxxD_KYI4YTH2W.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1 full </h3>
<div class="score-percentage">Score: 38.80%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/dwlBjf5FWq_r79tkh-OeL.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux 1 Pro </h3>
<div class="score-percentage">Score: 61.20%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/FxxC03lpwIy-GhkdF-Q8x.jpeg" width=500>
</div>
</div>
</div>
</div>
## About Rapidata
Rapidata's technology makes collecting human feedback at scale faster and more accessible than ever before. Visit [rapidata.ai](https://www.rapidata.ai/) to learn more about how we're revolutionizing human feedback collection for AI development.
# Rapidata Hidream I1全量偏好数据集
<a href="https://www.rapidata.ai">
<img src="https://cdn-uploads.huggingface.co/production/uploads/66f5624c42b853e73e0738eb/jfxR79bOztqaC6_yNNnGU.jpeg" width="400" alt="数据集可视化">
</a>
本文本到图像（Text-to-Image，T2I）数据集包含来自超3.8万名独立标注者的逾19.5万条人类标注反馈，仅用约1天时间便通过[Rapidata Python API](https://docs.rapidata.ai)完成采集，全平台开放访问，非常适用于大规模模型评估。
本次评估围绕偏好性、连贯性与对齐性三大维度展开。
可前往我们的[官方网站](https://www.rapidata.ai/benchmark)查看最新的模型排名榜单。
若您从本数据集获益并希望后续获取更多同类资源，欢迎点赞支持❤️
## 数据集概览
本T2I数据集包含来自超3.8万名独立标注者的逾19.5万条人类标注反馈，采集周期仅约1天。本次评估围绕偏好性、连贯性与对齐性三大维度对Hidream I1全量模型展开。
本次评估采用1v1对比模式，将Hidream I1全量模型与其余14款模型进行对比，分别为：Halfmoon-4-4-2025、OpenAI 4o-26-3-25、Ideogram V2、Recraft V2、Lumina-15-2-25、Frames-23-1-25、Imagen-3、Flux-1.1-pro、Flux-1-pro、DALL-E 3、Midjourney-5.2、Stable Diffusion 3、Aurora以及Janus-7b。
> **注：** 模型名称后的数字（如Halfmoon-4-4-2025）代表图像生成日期（即2025年4月4日），用于说明所使用的模型版本。
## 对齐性维度
对齐性评分用于量化生成图像与输入提示词的匹配程度。标注者被问及："哪张图像更贴合描述文本？"
<div class="vertical-container">
<div class="container">
<div class="text-center">
<q>车底的人</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/w3pNtIgrKMMULW77oBqJ4.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux-1.1-pro</h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/0g6ipnW8J4PX0VXnIWK2k.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>猫背上的椅子</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/OA9nYfpfewJG98iJ-3G2Y.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">OpenAI 4o-26-3-25</h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/MMOdg5L4cJW7wSLxT-QO0.jpeg" width=500>
</div>
</div>
</div>
</div>
## 连贯性维度
连贯性评分用于衡量生成图像是否具备逻辑一致性，且无人工痕迹或视觉瑕疵。在不查看原始提示词的前提下，标注者被问及："哪张图像存在**更多**瑕疵，且**更有可能**是AI生成作品？"
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">瑕疵占比：5.84%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/GSqTj1Ole9l9UaM1cnNLc.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Janus-7B</h3>
<div class="score-percentage">瑕疵占比：94.16%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/v2b-vDkUG1Y6paClTOFuK.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">瑕疵占比：92.46%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/y5DcMgqhxSqH3mbKaIDJP.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Imagen-3</h3>
<div class="score-percentage">瑕疵占比：7.54%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/REveLQYhVXC51gWEM7YU6.jpeg" width=500>
</div>
</div>
</div>
</div>
## 偏好性维度
偏好性评分用于反映参与者对单张图像的视觉喜爱程度，不受提示词约束。标注者被问及："你更偏好哪张图像？"
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">得分：70.04%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/hH1ewWQ4tYtqaiSd7qZF7.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Stable Diffusion 3</h3>
<div class="score-percentage">得分：29.96%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/Uu5pl6zBxxD_KYI4YTH2W.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Hidream I1全量模型</h3>
<div class="score-percentage">得分：38.80%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/dwlBjf5FWq_r79tkh-OeL.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux 1 Pro</h3>
<div class="score-percentage">得分：61.20%</div>
<img style="border: 3px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/664dcc6296d813a7e15e170e/FxxC03lpwIy-GhkdF-Q8x.jpeg" width=500>
</div>
</div>
</div>
</div>
## 关于Rapidata
Rapidata的技术让大规模人类反馈采集工作变得比以往任何时候都更快捷、更易获取。欢迎访问[rapidata.ai](https://www.rapidata.ai/)，了解我们如何革新AI开发领域的人类反馈采集流程。
提供机构：
maas
创建时间：
2025-05-17
