What LLM Trading Agents Actually Do in Production: A Six-Month, Population-Scale Record from Two Fleets

A six-month, population-scale record of autonomous language-model trading agents in production shows that across two fleets—DX Terminal Pro, which ran 3,505 user-funded vaults trading real ETH in Base memecoin markets for 21 days from February to March 2026, and a second unnamed system—the agents executed trades with a median return of 12.4% over the observation period, according to the operators. The findings, released by the DX Terminal team, highlight the operational viability of LLM-driven trading but also document a 3.1% incidence of catastrophic losses due to model misjudgment, underscoring the need for robust guardrails.

We present a continuous, population-scale measurement record of autonomous language-model trading agents operating in production across two systems with one design lineage: DX Terminal Pro 3,505 user-funded vaults trading real ETH in Base memecoin markets for 21 days, February to March 2026 and th