Open Models Prompt Dataset with 5400+ Image Prompts and Generated Images

A collection of over 5400 prompts and their corresponding generated images for open-source AI image models, exceeding 10GB in total size. The dataset was created by Goku-OpenLab and was last updated on July 21, 2026. It includes full metadata and preview images for each prompt.

Use Cases

• Training or fine-tuning text-to-image models based on the provided prompt-image pairs.
• Analyzing the relationship between prompt text and generated image characteristics based on the included metadata.
• Benchmarking the performance of different open-source image generation models using the structured prompt collection.
• Studying prompt engineering techniques by examining the variety of prompts and their visual outcomes.

Strengths

• Contains over 5400 prompt-image pairs, providing a substantial corpus for analysis.
• Exceeds 10GB in total data volume, indicating significant content.
• Includes full metadata and preview images for each entry, suggesting detailed annotations.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-07-21 05:28:56; freshness should be verified.

Provenance

Source

Goku-OpenLab on Hugging Face.

Collection Method

Likely a curated collection of prompts and images generated using various open-source AI image models.

Freshness

Last updated July 21, 2026.