1
Long-form financial reasoning · risk detection
Strong on financial-document reasoning
US · A-shares · HKLive preview · 6 models running100% Free ForeverFrontier Model Ensemble · 2026
100% Free ForeverFree daily reports · No credit card · Unsubscribe anytime
Pick your stocks.
Every trading day,
get a
Wall-Street-grade AI research report.
StockKit runs Claude Fable 5 · GPT-5.6 Sol · DeepSeek-V4 Pro · Gemini 3.5 Flash · Grok 4.3 · Qwen3.7-Max — six of the world's top frontier models reasoning in parallel, fused with 32 real-time data sources, generating an institutional-grade report for your 1-3 subscribed stocks at the time you choose, straight to your inbox. Completely free, forever, unsubscribe anytime.
Get free daily reports
Choose a market + 1-3 stocks · delivered at the time you set
Live preview · 6 models runningSample report
AAPL · Apple Inc.
Composite score
72 / 100
Key support
$208.40
Key resistance
$224.10
Data confidenceHigh · 5/6 consensus
Sample data for illustration only — not investment advice
Frontier Model Ensemble · 2026
Six frontier models running at once — a multi-model approach few tools take
Most products use one model at most. StockKit calls Claude Fable 5, GPT-5.6 Sol, DeepSeek-V4 Pro, Gemini 3.5 Flash, Grok 4.3, Qwen3.7-Max — each the latest flagship with its own strengths, generating independently, then cross-checked sentence by sentence by an arbitration layer. Any one model's hallucination is corrected on the spot by the other five.
2
Multi-step reasoning · data synthesis
Strong general reasoning and data synthesis
3
Quant modeling · mathematical reasoning
AIME math SOTA · top pick for quant finance
4
Real-time multimodal · chart parsing
Native understanding of candlestick charts / earnings PDFs
5
Real-time social · news ingestion
Native access to live financial sentiment on X
6
A-share / HK semantic understanding
SOTA on Chinese filings and disclosure text
Cross-model arbitration
When models disagree on the same conclusion, the arbitration layer downgrades the statement, flags the divergence, and shows it transparently as "Low confidence" — never forcing a conclusion.
How it works
From signup to report in just 4 steps
No black box, no magic. Subscribe once, receive automatically every trading day — institutional-grade research right in your inbox.
Choose your market and 1-3 stocks
Covering 12,000+ tickers across US, A-shares, and HK. Pick up to 3 stocks to follow, leave your email, choose a delivery time (pre-market / after close / custom timezone) — all in under 30 seconds.
Supports Chinese/English company names, fuzzy search, sector filters
We pull 32 real-time data sources for you daily
Before each delivery, we automatically pull every latest data point for your subscribed stocks from 32 first-hand sources: exchange Level-2 quotes, SEC EDGAR · CNINFO · HKEX filings, Bloomberg · Refinitiv feeds, Reuters · Benzinga · CNBC news, Trefis valuation, X social sentiment, options chains, dark-pool prints, and more.
100% licensed / public APIs · faulty sources auto-downgraded · zero manual intervention
Six frontier models reason in parallel + arbitration
Claude Fable 5, GPT-5.6 Sol, DeepSeek-V4 Pro, Gemini 3.5 Flash, Grok 4.3, and Qwen3.7-Max independently analyze the same raw data; the arbitration layer compares them sentence by sentence, keeping ≥5/6 consensus, flagging divergence, and dropping low-confidence content.
≥5/6 consensus = High · 3-4/6 = Medium · ≤2/6 dropped
Get an institutional-grade report in your inbox at your chosen time
At your chosen time, the report arrives on schedule. 21 sections cover the executive summary, technical structure, valuation matrix, news catalysts, Base/Upside/Downside scenarios, a nine-dimension risk matrix, derivatives pricing, and a 1-5 trading-day watch checklist.
Free forever · no watermark · urgent push on major events · one-click unsubscribe
Inside every report
Every free report is packed with 12 institutional-grade capabilities
These cost $24,000/year on a Bloomberg Terminal. On StockKit they're always $0.
Full technical indicator suite
24 core indicators including MA5/20/60/120, RSI, MACD, KDJ, Bollinger Bands, ATR, OBV, and CCI — with AI auto-detecting trends, momentum, divergences, and pattern breakouts.
Valuation matrix
Full-dimension valuation across PE/PB/PS/EV-EBITDA/PEG/DCF, cross-referenced on three axes: sector median, the stock's own historical percentile, and sell-side consensus.
News catalyst timeline
News / earnings / regulatory events from the last 90 days, sorted by relevance and time, with AI tagging Material / Context / Noise impact levels and estimating price linkage.
Sentiment radar
Aggregates 6 sentiment sources — X, Reddit, Xueqiu, StockTwits, Discord, and news sentiment — with 7-day sentiment-score trends and anomaly detection.
Key levels + triggers
Three support and three resistance levels, plus pivot zones and institutional cost basis — each level annotated with how it formed, a validity score, and a playbook for what happens on a break above/below.
Three scenarios + Monte Carlo
Base / Upside / Downside scenarios plus 10,000 Monte Carlo path simulations, giving price-range distributions and probability density for the next 5/10/20 days.
Nine-dimension risk matrix
Nine quantified dimensions — data risk, volatility, liquidity, trend failure, catalyst failure, macro events, policy, FX, and counterparty — telling you exactly where the landmines may be.
Derivatives view
One-click reads on options IV / Skew / Put-Call Ratio / Max Pain / Gamma Exposure, so you can see what the big players are really positioning for.
Data confidence scoring
Every conclusion is tagged with its consensus level (≥5/6 High, 3-4/6 Medium, ≤2/6 Low); low-confidence content is auto-downgraded or dropped — the AI never makes things up for you.
Smart alert push
When your watchlist hits a key-level break, catalyst, risk event, or confidence reversal, you get millisecond Email / Webhook / Telegram alerts.
Three UI languages
Currently supports 3 UI languages: Simplified Chinese, Traditional Chinese, English. Japanese, Korean, and Spanish report generation stays out of public claims until it is visible in product.
Export / share / API
Export to PDF / Notion / Markdown / CSV, share via login-free links, and developers get a free REST + Webhook API.
Compare
Capabilities that used to cost $24,000/year on a terminal — now free, forever
Engineering optimization and shared inference keep the cost per report very low, which is how we can offer the free plan. The optional Plus plan and future enterprise offerings are how we earn revenue.
Capability
StockKit · Free
Bloomberg Terminal
Paid AI research site
Basic quote app
AI models
Six flagships in parallel · arbitration removes hallucination
No native AI
Single model · prone to hallucination
No AI
Data sources
32 first-hand real-time sources
200+ but terminal-only
3-5
1-2
Technical indicators
24 + pattern recognition
Complete
8-10
Requires paid plugin
Valuation models
PE/PB/PS/EV-EBITDA/PEG/DCF
Complete
PE only
None
News catalysts
90 days · 3-tier impact + price linkage
Real-time but no AI summary
Headline summary
None
Sentiment radar
6 sources + anomaly detection
Twitter only
None
None
Scenario analysis
Base / Up / Down + 10k Monte Carlo
Manual modeling required
None
None
Risk matrix
9-dim quantified + trigger playbook
Interpret it yourself
Generic disclaimer
None
Derivatives view
Options IV / Skew / Gamma / Max Pain
Complete
None
None
Confidence
Per-sentence tagging · low-confidence dropped
None
None · black-box output
None
Daily updates
Free · real-time incremental
Build it yourself
$29 / mo
No customization
Export / API
Free · no watermark · public API
Limited · expensive
$49 / mo · capped
Watermarked
Languages
3 UI languages: Simplified Chinese, Traditional Chinese, English
Mostly English
Single
Single
Price
$0 forever · unlimited
$24,000 / year
From $199 / mo
Free · limited features
Fair by design
An AI that's fair to every user
Retail users have long been shut out by paywalls, stock-tip groups, and conflicted sell-side reports. We flip that — putting an institutional research framework into every ordinary user's hands for free, with the same rules for everyone.
No insider routingNo paid placementsNo buy/sell calls
Fully transparent algorithm
Every indicator is annotated with its formula and data source; every AI conclusion shows which models agreed.
Six-model arbitration, no single-point bias
Any single model's hallucination is corrected on the spot by the other five independent models; anything below 3/6 confidence is dropped. We take no payment from listed companies, brokers, or market makers for content or ranking.
No selling, no tips, no nudging
StockKit accepts no broker kickbacks, promotes no specific stock, and gives no buy/sell advice. We build research tools, not advisory services.
Data privacy
Your watchlist, email, and viewing history are never sold to third parties.
Built for individual investors
What real users say
Retail investor · 5y
I used to spend an hour across 5 sites to piece together one stock's picture — now it's 7 seconds, and the data granularity beats Yahoo.
ETF blogger
Running multiple models at once really sets it apart. Some tools surface unverified earnings numbers; StockKit flags Low confidence and tells me to verify.
Part-time trader
Completely free but nothing like a free tool. The PDF layout is cleaner than the paid service I used to subscribe to.
Daily updates · still free
Save 1-3 stocks, get updates at the time you set
At your chosen time, we push the latest report changes, key-level moves, and news catalysts for your watchlist to your inbox. Completely free, unsubscribe anytime.
Verify your email once, use forever
Timezone options · pre-market / after close / custom
Major changes get an ⚠️ urgent push
you@domain.com
Market
US stocks
Delivery time
Pre-market
Stocks · up to 3
add ticker
Email verification required before sending · one-click unsubscribe · free forever
Generated in ~7 seconds
Pick up to 3 stocks you care about most and get a report every trading day
Just your email · no credit card · free forever · one-click unsubscribe
Get free daily reportsSubscribed by 1,820,000+ investors · ⭐ 4.9 / 5