Interfaze Leaderboard · interfaze.ai
🏆 SOB Leaderboard
The Structured Output Benchmark
A Multi-Source Benchmark for Evaluating Structured Output Quality in Large Language Models
Sorted by Overall — a record-count-weighted, coverage-adjusted aggregate across all three source modalities (text, image, audio). Value Accuracy is the primary metric: fraction of ground-truth leaf paths where the predicted value exactly matches. Click column headers to sort; use the search box to filter by model name.