AI-powered stock picking
Compare the stock picking capabilities of leading AI models through a common process, shared dataset, and quarterly rebalance.
Disclaimer: The IA 500 and other Intelligent Alpha stock picking benchmarks presented are experimental, AI-powered financial benchmarks developed by Intelligent Alpha for educational and informational purposes only. The IA 500 and other Intelligent Alpha stock picking benchmarks are not a financial product, investment vehicle, index fund, or investable strategy. It does not represent a recommendation to buy, sell, or hold any security. No assets are managed according to the IA 500 or other Intelligent Alpha stock picking benchmarks, and it should not be used as the sole basis for any investment decision. The benchmark reflects hypothetical, backtested, and/or forward-looking results generated by AI models. The IA 500 and other Intelligent Alpha stock picking benchmarks do not reflect actual portfolio performance, real-world trading, transaction costs, slippage, taxes, or other factors that would affect returns in practice. Hypothetical results have inherent limitations and should not be interpreted as indicative of future performance.
Performance
Select any model or ensemble from the models section to re-base the chart, metrics and returns.
Standard index returns are presented net of an assumed 0.25% annual management fee. The gross, no-fee series remains available as a secondary published data series.
Returns by period
Click a period to scope the model ranking.
* Periods marked with an asterisk are backtested via models with knowledge cutoff dates outside the date window.
MODELS
Top holdings for each model, and how each model family ranks in excess return.
Overlap
Pick two portfolios to see how their top 25 stock picks agree and diverge.
Methodology
IA 500 Benchmark Methodology · Effective August 6, 2026
At the start of each quarter, the most advanced available model for each model family runs the same stock picking process on a large cap stock universe. Models are given a common dataset of information about the universe including valuation metrics and forward estimates.
In addition to each individual model, we test the IA 500 as an ensemble of all model picks. In the ensemble, each model's portfolio is given an equal weight in the aggregate combination of all model portfolios.
All models start at a 1,000 base index value at inception. Each model's new stock picks and weights are applied at a review that happens at the beginning of each new quarter. Then holdings drift with market prices until the next review, with dividend cash reinvested at the index level.
Disclaimer: AI-generated predictions are inherently uncertain and may be inaccurate. Past performance, whether actual or simulated, is not indicative of future results. The models used to generate the IA 500 and other Intelligent Alpha stock picking benchmarks may change over time without notice, and Intelligent Alpha makes no representation or warranty regarding the accuracy, reliability, or completeness of any information provided.
The IA 500 and other Intelligent Alpha stock picking benchmarks use artificial intelligence tools to assist in the construction, evaluation, and documentation of model-driven equity portfolios. The AI tools are large language models and related analytical systems that process structured inputs such as company data, market information, portfolio constraints, ranking criteria, and benchmark methodology instructions. The tools are used to generate or support portfolio selections, qualitative assessments, and comparative analysis under a defined methodology.
The AI tools do not independently verify all underlying data and should not be understood as providing investment advice, fiduciary judgment, or a guarantee of future performance. Outputs depend on the quality, completeness, timeliness, and accuracy of the data and instructions provided. Results may be incorrect, incomplete, stale, or misleading if source data is inaccurate, if market conditions change rapidly, if relevant company information is unavailable or misinterpreted, if model assumptions are inappropriate, or if the AI system makes reasoning, classification, or calculation errors.
The IA 500 and other Intelligent Alpha stock picking benchmark methodologies include human oversight, defined portfolio construction rules, and review procedures intended to reduce these risks, but such controls cannot eliminate them. AI-generated outputs may reflect limitations in training data, model design, prompt construction, data availability, or benchmark implementation. The resulting portfolios and benchmark results should therefore be evaluated as model-assisted analytical outputs, not predictions, recommendations, or assurances of investment performance.
Intelligent Alpha's stock-related benchmarks may have different inception dates or start periods, may apply different fees or hypothetical fee assumptions as determined by Intelligent Alpha, and may be relevant to different traditional benchmarks based on their respective objectives and universes. Intelligent Alpha makes best efforts to make its benchmarks comparable with one another and, where relevant, with external benchmarks; however, these differences may limit direct comparisons.
Nothing on this page constitutes investment advice, tax advice, legal advice, or a solicitation to invest. Users should consult a qualified financial professional before making investment decisions. By accessing this content, you acknowledge that you understand these limitations and agree not to rely on the IA 500 and other Intelligent Alpha stock picking benchmarks for investment purposes.