BI-Bench (Python, 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; Python setting with the BI-Agent search, transform and join tools added to the code-execution tool). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O4 Mini | 66.3 |
| 2 | GPT-5.5 | 65.4 |
| 3 | Kimi K2.6 | 62.7 |
| 4 | GPT-5.2 | 60.2 |
| 5 | DeepSeek V4 Pro | 57.6 |
| 6 | Mistral Large 3 | 51.1 |
| 7 | GPT-4o | 44.7 |
| 8 | Llama 4 Maverick | 38.8 |
| 9 | GPT-OSS-120B | 27.3 |
| 10 | Qwen 3 8B | 13.1 |
Interactive version: theaggregate.ai/benchmark?slug=bi-bench-python-bi-agent-tools · How It Works · Data refreshed daily, snapshot 2026-09-26.