Seedream-3_t2i_human_preference
收藏魔搭社区2026-01-06 更新2025-08-16 收录
下载链接：
https://modelscope.cn/datasets/Rapidata/Seedream-3_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;
}
.horizontal-container {
display: flex;
flex-direction: row;
justify-content: center;
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 img.big {
max-height: 350px; /* Set the desired height */
}
.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 Seedream 3 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 ~400'000 human responses from over ~30'000 individual annotators, collected in less than 7h using the [Rapidata Python API](https://docs.rapidata.ai), accessible to anyone and ideal for large scale evaluation.
Evaluating OpenAI 4o (version from 26.3.2025) 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 ❤️
To add your own model to the benchmark send us an e-mail at: jason@rapidata.ai
## Overview
The evaluation consists of 1v1 comparisons between Seedream 3 (version from 24.7.2025) and 16 other models:
- 4o
- Flux-1-pro
- Flux-1.1-pro
- imagen 4 ultra
- Aurora
- Imagen-3
- DALL-E 3
- Midjourney-5.2
- Frames-23-1-25
- Stable Diffusion 3
- Janus-7b.
- hidream-l1-full
- Recraft V2
- Ideogram V2
- halfmoon-4-4-25
- Lumina-15-2-25
## 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 black colored banana.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/EvMP6G5tk_x4W1gYGK_zb.jpeg" width=500>
</div>
<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/672b7d79fd1e92e3c3567435/zYrtQZQr42elPG3AT9USw.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>A pizza on the right of a suitcase.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">Score: 0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/k8VK18hbNJ0gu4C_9UOZS.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Imagen 4 Ultra </h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/JCEWcyUhfqCmxjklYeTqW.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">Seedream 3 </h3>
<div class="score-percentage">Glitch Rating: 0%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/h4QyNuxUJxCd593ECZm8c.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Janus 7b </h3>
<div class="score-percentage">Glitch Rating: 100%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/m0prttWSwiIgwt-akumPh.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">Glitch Rating: 100%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/X5K-zAABjbfeAt88S2FK4.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Ideogram </h3>
<div class="score-percentage">Glitch Rating: 0%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/B-bun8BPmd-Y7bDxsc48d.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">Seedream 3 </h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/ZLpBia3tQpjHkVXh0Ksvq.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Lumina-15-2-25</h3>
<div class="score-percentage">Score: 0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/WwGrkiRQhvKy38Ilpk6bM.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">Score: 0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/H86Vrl06r8vapiW60Aq6A.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux 1 Pro </h3>
<div class="score-percentage">Score: 100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/s4B76J8qfj-3zx7a71URO.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 Seedream 3 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="Dataset visualization">
</a>
该文本到图像（Text-to-Image, T2I）数据集包含来自约3万名独立标注员的超40万条人类标注反馈，通过[Rapidata Python API](https://docs.rapidata.ai)在不足7小时内完成采集，全量开放且适配大规模模型评估场景。
本次评估针对OpenAI 4o（2025年3月26日版本），从偏好性、连贯性与对齐性三个维度展开。
可访问我们的[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❤️
若希望将您的模型加入基准测试榜单，请发送邮件至：jason@rapidata.ai
## 数据集概览
本次评估包含Seedream 3（2025年7月24日版本）与其余16款模型的一对一对比测试，参与对比的模型如下：
- 4o
- Flux-1-pro
- Flux-1.1-pro
- imagen 4 ultra
- Aurora
- Imagen-3
- DALL-E 3
- Midjourney-5.2
- Frames-23-1-25
- Stable Diffusion 3
- Janus-7b
- hidream-l1-full
- Recraft V2
- Ideogram V2
- halfmoon-4-4-25
- Lumina-15-2-25
## 对齐性维度
对齐性评分用于衡量生成图像与输入提示词的匹配程度。标注任务为：「哪张图像更贴合给定描述？」
<div class="vertical-container">
<div class="container">
<div class="text-center">
<q>一根黑色的香蕉。</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/EvMP6G5tk_x4W1gYGK_zb.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Hidream-L1 Full </h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/zYrtQZQr42elPG3AT9USw.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>行李箱右侧的披萨。</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/k8VK18hbNJ0gu4C_9UOZS.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Imagen 4 Ultra </h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/JCEWcyUhfqCmxjklYeTqW.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的逻辑自洽性，以及是否存在伪影或视觉瑕疵。本次任务不向标注员展示原始提示词，提问为：「哪张图像的瑕疵更多，且更像是AI生成的？」
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">瑕疵评分：0%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/h4QyNuxUJxCd593ECZm8c.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Janus 7b </h3>
<div class="score-percentage">瑕疵评分：100%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/m0prttWSwiIgwt-akumPh.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">瑕疵评分：100%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/X5K-zAABjbfeAt88S2FK4.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Ideogram </h3>
<div class="score-percentage">瑕疵评分：0%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/B-bun8BPmd-Y7bDxsc48d.jpeg" width=500>
</div>
</div>
</div>
</div>
## 偏好性维度
偏好性评分用于衡量标注员对图像视觉效果的喜爱程度，不受提示词约束。标注任务为：「你更偏好哪张图像？」
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/ZLpBia3tQpjHkVXh0Ksvq.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Lumina-15-2-25</h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/WwGrkiRQhvKy38Ilpk6bM.jpeg" width=500>
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Seedream 3 </h3>
<div class="score-percentage">得分：0%</div>
<img src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/H86Vrl06r8vapiW60Aq6A.jpeg" width=500>
</div>
<div>
<h3 class="score-amount">Flux 1 Pro </h3>
<div class="score-percentage">得分：100%</div>
<img style="border: 5px solid #18c54f;" src="https://cdn-uploads.huggingface.co/production/uploads/672b7d79fd1e92e3c3567435/s4B76J8qfj-3zx7a71URO.jpeg" width=500>
</div>
</div>
</div>
</div>
## 关于Rapidata
Rapidata的技术可实现大规模人类反馈的高效采集，相较以往方案更为快捷易用。访问[rapidata.ai](https://www.rapidata.ai/)了解更多关于我们如何革新AI开发中的人类反馈采集技术的详情。
提供机构：
maas
创建时间：
2025-08-13
