ESQ-Bench (Tier 3): leaderboard

Metric: Execution match (%; among executed queries; Tier-3 schemas; over ESQ-Bench's Oracle-first enterprise NL2SQL question bank; execution match among executed queries; schema-linked prompt, temperature 0). Source: arxiv.org. Saturation forecast: Around December 2026. 3 models tracked.

Top models

#ModelScore
1Claude Sonnet 4.668.7
2GPT-4o57.2

Interactive version: theaggregate.ai/benchmark?slug=esq-bench-tier-3 · How It Works · Data refreshed daily, snapshot 2026-09-26.