Together AI· Inference platform

AI for Systems: Using LLMs to Optimize Database Query Execution

Summary We worked in collaboration with Stanford University, the University of Wisconsin–Madison, and Bauplan to test whether LLMs can optimize database query execution plans. The results show that LLM-guided plan rewrites can improve execution performance without modifying the database engine itself. Recent advances in AI have been driven by improvements in the underlying systems infrastructure. But this relationship does not have to be asymmetric: AI and LLMs can also be used to optimize the functional components of large-scale systems themselves. Our recent paper shows how AI can be used for database query optimization. Traditional query optimization relies on cost-based estimators that calculate execution paths using statistical models and predefined heuristics. When executing a query like "find all sci-fi shows with space travel," …

Announced

03 Apr 2026 00:00 UTC