How people actually prompt
generative AI
1,000+ real prompts from viral posts on X — classified by model, technique, theme, and reference type. Every entry traces back to a practitioner who shipped something worth sharing.
A prompt observatory. Zero synthetic data. Zero crowdworkers. Just the organic distribution of how people talk to image and video models.
How it's built
No crowdworkers. No GPT-generated prompts. Every entry traces back to a real post by a real practitioner.
Collect
Viral image and video generation posts are sourced from X. Only posts with high engagement pass the filter — likes, reposts, and replies act as a quality signal.
Extract
Claude strips social framing ("check out this image I made!") to isolate the raw generation prompt. The model, technique, and reference type are detected automatically.
Label
Each prompt is classified across 5 dimensions: category, visual theme, art style, reference requirements, and model family. Every label links back to its source post.
Image vs. Video
The split between image generation and video generation prompts in the dataset.
Related prompt datasets
Several prompt datasets exist for evaluating generative AI models. This collection serves a different purpose — documenting organic practitioner behavior rather than providing evaluation prompts.
| Dataset | Size | Source | Modality | Engagement | Curated |
|---|---|---|---|---|---|
| DrawBench | 200 | Synthetic (LLM) | Image | None | 2022 |
| PartiPrompts | 1,632 | Crowdworkers (Google) | Image | None | 2022 |
| T2I-CompBench | 6,000 | Synthetic (GPT-4) | Image | None | 2023 |
| GenAI-Bench | 1,200 | LLM + human mix | Image + Video | None | 2024 |
| EvalCrafter | 700 | LLM + real users | Video | None | 2024 |
| VBench | 1,600 | Manual per dimension | Video | None | 2024 |
| T2VEval-Bench | 1,783 | LLM + manual | Video | Lab MOS | 2025 |
| ummerr/prompts(this dataset) | - | Organic / in-the-wild | Image + Video | Viral filter | Mar 2026 |
Start exploring
Filter by model, technique, or theme. Copy what works. See what the community is actually making.
Browse prompts →