Impressionist Landscape Paintings for AI Training, 4,308 Images

4,308 Impressionist landscape paintings curated from the OpenBrush-75K collection. The subset was created by the author jaddai to provide a targeted dataset for training AI models, specifically LoRAs. It was last updated on May 27, 2026.

Use Cases

• Train a LoRA for Impressionist landscape generation based on the described style and genre combination.
• Fine-tune a text-to-image model on a curated set of landscape paintings based on the described caption schema.
• Benchmark the performance of vision-language models on art style classification based on the described VLM annotations.

Strengths

• Contains 4,308 specifically curated images, allowing users to avoid downloading the full 75,313-image parent dataset.
• Maintains the same CC0 license and caption schema as the parent dataset, ensuring consistency and permissive use.
• Annotations were generated using a specific vision-language model (Qwen3-VL-30B-A3B), suggesting a degree of automated consistency.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known (4,308), but the total size, file formats, and sample data are unknown, which may limit suitability assessment.
• Data may reflect the selection and annotation bias inherent to the parent OpenBrush-75K collection and the specific VLM used.

Provenance

Source

Curated subset of the jaddai/openbrush-75k dataset on Hugging Face.

Collection Method

Cross-cut selection based on style (Impressionist) and genre (landscape) from a larger collection.

Freshness

Last updated 2026-05-27 21:24:08; freshness should be verified.