Using LLMs to Find Alphas: Grok Surprisingly Outperforms in Quant Backtests

SamuelAlbanie · x · 2026-08-11

A researcher shared an experiment using Large Language Models to discover Alpha factors in quantitative trading. The LLMs were tasked with building technical features and given backtest models to evaluate their performance.

Comparing metrics like the Sharpe ratio against token usage and cost, Grok delivered surprisingly strong results on these real-world financial tasks. This suggests that xAI may prioritize practical task-solving capabilities over simply benchmarking well.

Original post →

More from Research

Research channel →