Prompt Dataset: Text-to-Image Prompts and Captions for AI Training

Quarry Prompt Datasets is a collection of LanceDB tables containing text-to-image prompts and captions imported from public datasets on the Hugging Face Hub. The dataset is authored by jtreminio and was last updated on June 15, 2026. It is designed to be consumed by the SwarmUI-Quarry extension for indexing and retrieval.

Use Cases

• Training or fine-tuning text-to-image generation models based on the provided prompts.
• Benchmarking model performance using pre-existing prompt and caption collections.
• Indexing and retrieving prompts for creative AI applications via the SwarmUI-Quarry extension.
• Analyzing trends and patterns in text-to-image prompt engineering.

Strengths

• Contains text-to-image prompts and captions, which are core features for generative AI training.
• Data is stored in LanceDB tables, a format designed for efficient vector search and retrieval.
• Last updated on June 15, 2026, suggesting recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Aggregated from publicly available datasets on the Hugging Face Hub.

Collection Method

Imported and structured into LanceDB tables.

Freshness

Last updated 2026-06-15 16:04:29; freshness should be verified.