Exploring Autonomous Agentic Data Engineering: leaderboard
Large Language Models (LLMs) have demonstrated strong performance on general tasks, while often struggling to adapt to specialized domains without high-quality domain-specific data. Existing LLM-based data curation methods primarily rely on human-designed workflows, leaving it unexamined whether LLMs can autonomously execute an end-to-end data engineering pipeline for model specialization. We form
Metric: MATS Science (Scratch) (self-reported). Source: benchmarklist.com. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3.1 | 7.13 | #264 |
| 2 | Claude Sonnet 4 | 3 | #199 |
| 3 | DeepSeek R1 | 2.5 | #251 |
| 4 | Gemini 2.5 Pro | 2 | #152 |
| 5 | GPT-5.2 | 2 | #110 |
| 6 | Qwen 3 Max | 1.58 | #208 |
Interactive version: theaggregate.ai/benchmark?slug=exploring-autonomous-agentic-data-engineering · How It Works · Data refreshed daily, snapshot 2026-10-20.