Project record01 mention
Originally shared on Hacker News
Vistarium
This web app uses a local AI model (Qwen 3.8 27B) to evaluate and tag public domain images from US national parks to find the best landscapes. Images are sorted by predicted aesthetic rating and can be filtered by time of day, color, orientation, and park.
01 / Source history
Shared over time
Each entry reflects what was shared on that date. Details may have changed since then.
Vistarium
This web app uses a local AI model (Qwen 3.8 27B) to evaluate and tag public domain images from US national parks to find the best landscapes. Images are sorted by predicted aesthetic rating and can be filtered by time of day, color, orientation, and park.
Posted by @pivot_root · Original comment ↗
“Using a local model (qwen 3.8 27b) to evaluate and tag public domain images from US national parks to find the best landscapes. They are sorted by predicted aesthetic rating and can be filtered by time of day, color, orientation, and park.”