BI-Bench (SQL, BI-Agent Tools): leaderboard

Metric: Accuracy (%; 100 end-to-end business-intelligence queries curated from real Power BI dashboards (82 projects, raw source tables needing search, reshaping, joins and analysis); an answer is correct when the output table matches the ground-truth table in shape and cell values up to row and column order, with numeric tolerance; mean over 10 runs; SQL setting with the BI-Agent search, transform and join tools added to the code-execution tool). Source: arxiv.org. Saturation forecast: Around November 2027. 10 models tracked.

Top models

#ModelScore
1O4 Mini61.9
2GPT-5.560.2
3GPT-5.255.2
4GPT-OSS-120B45.3
5Mistral Large 344.8
6GPT-4o37.7
7Kimi K2.636.7
8Llama 4 Maverick31.2
9DeepSeek V4 Pro30.8
10Qwen 3 8B19.8

Interactive version: theaggregate.ai/benchmark?slug=bi-bench-sql-bi-agent-tools · How It Works · Data refreshed daily, snapshot 2026-09-26.