AI-Powered Market
Intelligence for Every Artwork

Per-Artwork AI Valuation

Every artwork in the database receives an individual estimate — not an artist average, not a segment aggregate. An estimate for this specific work, based on its unique attributes: artist, medium, size, creation year, provenance signals, and market conditions at the time of calculation.

Confidence Ranges

Every estimate includes min/mid/max values with a confidence interval (e.g., $28.4M ±8.5%). You see how certain the model is, not just the number. Wider ranges mean more uncertainty — the model tells you when it's less confident, not less than it knows.

100+ Model Features

The XGBoost model considers over 100 features per artwork: artist career trajectory, medium, dimensions, creation period, provenance, auction house, sale season, lot position, price segment, artist momentum, and market conditions. Trained on 10M+ records across 1,400+ houses, back to 1986.

Ranked Comparable Sales

AI-generated comparable sets showing the 8–10 most relevant recent sales, ranked by similarity. Each comparable includes an explanation of why it was selected and how relevant it remains given recent market movement.

Methodology Transparency

Published methodology documentation. You know exactly how the estimate is calculated, what features drive it, and how the model is validated. Detailed validation reports shared with institutional clients under NDA.

API & Platform Access

LAE estimates are available on every artwork in the platform, the mobile app, and through the REST API. The current_estimated_price, current_estimated_price_min, and current_estimated_price_max fields appear on every artwork response.