DAXBench: leaderboard
Benchmark for evaluating how well language models write production-grade DAX for Power BI and Analysis Services business-intelligence scenarios.
Metric: Score (percentage). Source: www.daxbench.com. Source: benchmarklist.com. Saturation forecast: Around January 2027. 50 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 Mini | 96.2 | #212 |
| 2 | GPT-5.3 | 96.2 | #103 |
| 3 | GPT-5.3 Codex (High) | 88.6 | #69 (GPT-5.3 Codex) |
| 4 | GPT-5.5 | 86.7 | #31 |
| 5 | GPT-OSS-120B | 85.6 | #341 |
| 6 | GPT-5.1 Codex Max | 85 | #100 |
| 7 | Gemini 3 Flash (Preview) | 83.8 | #82 |
| 8 | GPT-5.4 (High) | 83.2 | #79 (GPT-5.4) |
| 9 | Gemini 3 Pro (Preview) | 81.3 | #66 |
| 10 | DeepSeek R1 | 81.3 | #251 |
| 11 | O3 | 80.2 | #124 |
| 12 | GPT-5.2 Instant | 80.1 | #210 |
| 13 | DeepSeek V3.2 | 79.9 | #205 |
| 14 | GPT-5.2 | 78.4 | #110 |
| 15 | Kimi K2 (Thinking) | 78.4 | #243 (Kimi K2) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=daxbench · How It Works · Data refreshed daily, snapshot 2026-10-20.