SchemaBridge Round Trip - SchemaBench Domain F1 Basic: leaderboard

SchemaBridge evaluates relational-schema design, description and reconstruction. Experiment-specific cohorts retain their own populations. Graph edit distance, hallucination, structural F1, text similarity and identifier coverage are distinct measurements. Round-trip runs use a fixed GPT-4o Mini constructor with each named describer. This board reports micro structural Domain F1 (0-1). Higher is better.

Metric: Micro structural Domain F1 (0-1). Source: arxiv.org. Saturation forecast: Around 2034. 5 models tracked. Updated 14 Nov 2026.

Top models

#ModelScoreOverall rank
1deepseek-coder-6.7B-instruct0.339#7426
2CodeLlama-13B-Instruct-hf0.241#9043
3Claude Sonnet 4.60.221#165
4GPT-5.4 Mini0.173#398
5Llama 3.1 8B Instruct0.167#4545

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=schemabridge-round-trip-schemabench-domain-f1-basic · How It Works · Data refreshed daily, snapshot 2026-11-16.