We used OpenAI's latest flagship reasoning model, o3-mini, paired with their newest product, Deep Research, to identify the most investable mid-cap stock in America.
We used OpenAI's o3-mini (high) in Deep Research mode to generate a stock recommendation for The GPT Investor. This essay outlines the technology, methodology, and findings.
📶BUY-$PSTG (Pure Storage Inc.) and hold till February 28th, 2026.
On January 31st, 2025, OpenAI released o3-mini—its newest reasoning model. According to OpenAI, o3-mini has "exceptional STEM capabilities—with particular strength in science, math, and coding—all while maintaining low cost and reduced latency."
Reasoning models introduce a new paradigm to Large Language Models by incorporating an internal "chain of thought" before responding to user queries. Similar to humans, the longer the model "thinks" about an answer, the better it performs on complex tasks. The o3-mini model is available in three variations that determine the amount of time it spends "thinking" before responding: low, medium, and high.
In the GPQA benchmark, designed by experts in biology, physics, and chemistry—where professionals with or pursuing PhDs in the respective fields achieve 65% accuracy—o3-mini (high) scored 79.7%.
Deep Research:
On February 2nd, 2025, OpenAI released Deep Research—an agent that uses reasoning to synthesize large amounts of online information and complete multi-step research tasks. It can conduct research independently by browsing the internet, collecting information, and reasoning over it to explore a topic set by the user.
According to OpenAI, "Deep Research is built for people who do intensive knowledge work in areas like finance, science, policy, and engineering and need thorough, precise, and reliable research."
In this essay, we will describe how we used o3-mini (high) in Deep Research mode to generate the latest stock recommendations.
Methodology:
We used the following prompt in the ChatGPT app with o3-mini (high) in Deep Research mode. This prompt outlines a structured process for identifying a mid-cap stock with the highest 12-month return potential.
It begins by selecting a promising sector, then systematically screens and analyzes three candidate companies based on recent financial data, growth catalysts, and valuation. The final selection is made by comparing each candidate’s projected return to the Russell Midcap Index and ranking them accordingly.
Throughout the process, the prompt emphasizes using a limited number of high-quality sources from the last 3–6 months to ensure an efficient, data-driven recommendation.