OpenAI shows GPT-5.6 Sol running quantum experiments through Codex
OpenAI describes how an MIT researcher uses GPT-5.6 Sol with Codex to run quantum computing experiments autonomously, analyze results and calibrate qubits.
OpenAI has published a case study on how its GPT-5.6 Sol model is being used in a physics lab through Codex. The example follows an MIT researcher who runs quantum computing experiments autonomously with the setup.
According to OpenAI, the workflow covers three tasks: launching experiments without manual supervision, analyzing the resulting data, and calibrating qubits. Qubit calibration is a repetitive and time-consuming routine on quantum hardware, and OpenAI says the model takes over that loop.
The published summary is a workflow account rather than a technical report. It does not specify the quantum hardware involved, how long the experiments ran, or any measurement figures. Readers who want to see where the model enters each step will find the details on OpenAI's page.
Full case study: OpenAI, How GPT-5.6 Sol helps run quantum computing experiments.
