SchemaBridge Description - RSchema BLEU4 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 bLEU4 (0-100). Higher is better.

Metric: BLEU4 (0-100). Source: arxiv.org. Saturation forecast: Around 2041. 5 models tracked. Updated 14 Nov 2026.

Top models

#ModelScoreOverall rank
1CodeLlama-13B-Instruct-hf7.63#9043
2Llama 3.1 8B Instruct6.37#4545
3Claude Sonnet 4.64.6#165
4GPT-5.4 Mini3.18#398
5deepseek-coder-6.7B-instruct2.9#7426

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

Interactive version: theaggregate.ai/benchmark?slug=schemabridge-description-rschema-bleu4-basic · How It Works · Data refreshed daily, snapshot 2026-11-16.