Most AI stock screeners promise the same thing: type in a ticker, get a "buy" rating, and supposedly watch the gains roll in. But when you open the hood, the magic disappears. The "AI" is a proprietary score you cannot inspect. The "buy" rating is trained on stale backtests. And the insider data is either missing or buried behind a paywall.
We built StoQuant for retail investors who are tired of that bait-and-switch. Our free stock screener starts with rules you can audit — Benjamin Graham value criteria — then layers in real insider-buying clusters, analyst actions, and quant coverage across 3,500+ small and mid-cap stocks. No black box. No hidden alpha claim. Just a deterministic filter engine plus the context you need to make your own call.
The phrase "AI stock screener" gets thrown around a lot. In practice it usually means one of three things:
None of these are necessarily bad, but they are rarely honest about what they can and cannot do. A real AI stock screener should tell you what signals it uses, how often those signals have historically worked, and — critically — when they fail.
That is why StoQuant publishes its own live accuracy proof. The composite Q-Score currently grades out at F: rank correlation near zero over 178 scoring dates, with a 46.1% beat rate against Russell 2000 over independent 90-day windows. We do not hide that number. For a stock screener, the most important feature is not a perfect score — it is a score you are allowed to disagree with.
A free stock screener usually has one of two business models. Either it shows you a handful of headline results and locks the real filters behind a paywall, or it sells your attention to advertisers and brokers. Neither serves the investor.
Paywalled filters create a strange incentive: the provider shows you just enough to make the paid tier look necessary, even when the free version should be useful on its own. Ad-supported screeners push you toward high-commission products or trendy stocks that generate clicks, not toward sound analysis.
StoQuant's free tier is designed to be the opposite. The full Graham filter set is available without a credit card. Insider cluster flags are included. The methodology and accuracy pages are public. The free screener is not a demo — it is a complete tool. The paid tiers add speed, customization, and power-user features, not basic functionality that should have been free from the start.
The core of our free screener is a deterministic Benjamin Graham filter set:
These are not secret. They come straight from Graham's Security Analysis and later writings on margin of safety. The point of using them is not that they predict returns — no rule set does that reliably — but that they narrow a 3,500-stock universe down to a small list of companies that pass a conservative quality-and-value test.
Because the rules are public, you can reproduce the screen yourself. You can argue with the thresholds. You can relax the debt filter or tighten the P/E cut. The screener is a starting point, not a prophecy.
Graham filters tell you what a company looks like on paper. Insider filings tell you what the people inside the company are doing with their own money.
StoQuant tracks roughly 300,000 insider transactions. A single purchase by a CEO is rarely meaningful on its own. But a cluster of purchases — multiple officers and directors buying within a short window — is a different signal. It suggests people with non-public operational visibility are voting with their wallets.
The catch is timing. A backtest that counts every insider buy as a signal looks great, because you can cherry-pick the ones that worked. An honest screener has to date each filing by when you could first act on it — Form 4 is due within two business days of a trade — and then track what happened next from that point. That is what our smart-money radar does, and it is included in the free screener results.
When you land on the StoQuant screener, you get:
You do not need a credit card to use the free tier. We do not spam your inbox. The goal is to give you a short, evidence-based watchlist every morning, then get out of your way.
For investors who want more, Pro adds unlimited custom screens, watchlists, and alerts. Power adds ML scores for the full 3,500-stock universe, portfolio optimization, and an API/MCP for agents. But the free screener is intentionally complete: if a free tool cannot stand on its own, the paid tiers are not worth paying for.
This is the part most marketing pages skip.
We do not claim the screener beats the market. Our own audit shows the composite Q-Score is currently indistinguishable from noise. We do not claim insider buying predicts returns — only that it is a piece of context worth knowing. We do not show backtested returns without explaining the methodology and the costs.
If you want a stock screener that promises 30% annual returns, StoQuant is not it. If you want a screener that tells you exactly what it is doing, shows you its report card, and lets you verify the output yourself, it probably is.
Start with the StoQuant stock screener and run the default Graham + insider screen. Read the methodology to see how the signals are built. Check the live accuracy proof to see the current grade. And if you want a daily digest of Q-Score changes and insider clusters, sign up for the free digest.
Performance figures refer to benchmark-relative outcomes vs the Russell 2000 over date-matched 90-day forward windows. Overlapping observations are thinned to independent windows before statistics are computed. Figures are signal-level and gross of execution costs unless stated. An earlier version of our accuracy page used mis-weighted benchmarks and fixed score bands labeled as quintiles; those figures were withdrawn on 2026-08-16 and replaced with the corrected measurement. See /proof for the live numbers.
Yes. The full Benjamin Graham filter set, insider cluster flags, the methodology pages, and the live accuracy dashboard are available without a credit card. The free screener is a complete tool, not a demo. Paid tiers (Pro and Power) add speed, customization, alerts, ML scores for the full universe, portfolio optimization, and API/MCP access.
A deterministic, public rule set: P/E below 15, P/B below 1.5, PEG below 1.0, debt-to-equity below 0.5, current ratio above 1.5, market cap between $250M and $10B, and analyst coverage below 8. The rules come straight from Graham's Security Analysis and his writings on margin of safety.
We do not claim that it does. Our own audit shows the composite Q-Score currently grades out at F — rank correlation near zero over 178 scoring dates and a 46.1% beat rate against the Russell 2000 over independent 90-day windows (see the methodology footnote in this post). We publish those numbers on the live accuracy page so you can disagree with the score on evidence, not marketing.
Daily Q-Score changes, insider buy clusters, and ML forecast updates for the tickers you follow. Email only, no account needed. See a sample digest.