Twelve Data

Implementation notes

When to use Twelve Data

Twelve Data offers the most generous free tier for financial market data in 2026: 800 calls/day, covering US and international stocks, ETFs, mutual funds, 100+ currency pairs, and major cryptocurrencies with historical OHLCV data going back decades. It's the best starting point for equity data projects — portfolio trackers, backtesting notebooks, stock screeners, financial dashboards — before you need real-time data or higher volume. The library ecosystem is solid (official SDKs for Python and Node) and the documentation is unusually clean for a finance API. Choose Twelve Data when you need a single provider for multiple asset classes; choose FRED for pure macro, Frankfurter for pure FX, and Binance direct endpoints for pure crypto if you don't need the aggregation.

Common pitfalls

• The 800 calls/day free limit sounds generous, but each call to the `time_series` endpoint counts as one call regardless of how many data points it returns. A 30-day daily OHLCV series for one symbol costs one call; fetching 50 symbols for a portfolio costs 50 calls.
• The `apikey=demo` parameter in examples only works for the specific symbol and interval shown in the documentation. Using `demo` with any other symbol returns a cached IBM or AAPL response, not an error — which can cause subtle bugs in code that checks only for HTTP errors.
• Real-time prices (`price` endpoint) return the current ask price for stocks when the market is open, and the most recent close when the market is closed. The response does not include a field indicating whether the market is currently open.
• International exchanges require the `:EXCHANGE` suffix on the symbol (e.g., `VODAFONE:LSE` for London Stock Exchange). Without it, the API defaults to the US market and either returns a 400 error or a different instrument entirely.
• The free tier has a per-minute rate limit of 8 requests/minute that's separate from the daily budget. A tight loop iterating over 50 tickers will hit the minute cap before the daily cap. Add a small delay (250-500ms) between calls or batch them via the multi-symbol endpoint syntax.

Quick start (bash)

curl "https://api.twelvedata.com/time_series?symbol=AAPL&interval=1day&outputsize=30&apikey=YOUR_KEY"

From our monitoring

Twelve Data has been one of the more reliable finance APIs in our monitoring — 99.7% uptime across the past 90 days with median response time around 350ms globally. The 429 rate-limit responses are honest and include a `Retry-After` header (rare among finance APIs). Data quality: US equities and major forex are audit-grade. Small-cap emerging-market stocks occasionally have data gaps (missing days, adjusted prices that don't match other sources). For anything hitting production, we'd cross-check against a second source for the specific tickers your users care about.

Production integration patterns

Portfolio tracker with delta refresh

The best pattern for a portfolio dashboard: at market open, fetch all watchlist symbols in one batch (Twelve Data supports comma-separated symbols in a single call — that's still counted as N calls, but the per-minute limit is more forgiving in batches). Cache the day's data. During the day, only refresh the top 5-10 most-viewed tickers on user interaction, not the whole watchlist. This turns a 50-symbol dashboard from 50 calls per refresh into 5-10, staying comfortably within the 800/day budget.

Screener with pre-computed universe

For stock screener apps (find all stocks with P/E < 20 and dividend yield > 3%), don't hit Twelve Data at query time — you'd blow through 800 calls in one user session. Instead, pre-compute the screening universe daily via a scheduled job that walks the S&P 500 or FTSE 100, stores fundamentals in your database, and serves queries from the database. Users get instant response, you use ~500 API calls per day, and the data is fresh enough for any non-day-trading use case.

Backtest data preparation pipeline

For backtesting notebooks: batch-download the full historical series for your test universe once, cache to Parquet or CSV, and iterate on the strategy against local files. Twelve Data's `time_series` with `outputsize=5000` returns ~20 years of daily data in one call. Download once, tune your strategy offline, only refresh when adding tickers or extending the date range. This is 100x cheaper than an ad-hoc 'call the API every time I re-run the notebook' pattern.

Compared to the closest paid alternatives

vs AlphaVantage — Free: 25/day · Premium: $50/month for 75 calls/minute

AlphaVantage's free tier is 25 requests/day — 32x tighter than Twelve Data. Its unique value is technical indicators as first-class endpoints (RSI, MACD, Bollinger Bands) with 200+ pre-computed indicators. If your app needs technical analysis and you don't want to compute indicators yourself, AlphaVantage's paid tier ($50/month) is easier to work with than Twelve Data. For raw price/OHLCV data, Twelve Data wins on quota and cleanliness.

vs Financial Modeling Prep — Free: 250/day · Starter: $19/month for 300 calls/minute

FMP is the best free source for fundamentals (balance sheets, income statements, DCF) — 250 calls/day. Twelve Data's fundamentals coverage is thinner and requires a paid tier for many endpoints. Use FMP for anything that involves reading financial statements or valuations; use Twelve Data for prices and time series. Many production apps use both.

vs Polygon.io — Free: limited sample · Starter: $99/month real-time

Polygon is the professional choice for real-time and tick-level equity data — think 'building an algo trading platform.' Free tier is basically a limited-endpoint sample; the useful tier starts at $99/month. Not relevant for hobbyist or personal projects, but worth knowing about if your app grows into real trading infrastructure. Twelve Data is the correct upgrade path first; Polygon comes after you've validated user demand.

Embed this badge

Add a live status badge to your README, docs, or website.

Markdown

[![Twelve Data Status](https://freeapi-builder.a10ayassine.workers.dev/badge/twelve-data.svg)](https://freeapi.watch/twelve-data)

HTML

<a href="https://freeapi.watch/twelve-data"><img src="https://freeapi-builder.a10ayassine.workers.dev/badge/twelve-data.svg" alt="Twelve Data Status"/></a>