Access point-in-time financial data derived directly from SEC EDGAR filings. Simple APIs designed for quant research and production fintech applications.
from openfindb import fundamentals
# Fetch point-in-time financials — no look-ahead bias
df = fundamentals.get(
ticker="AAPL",
fields=["revenue", "net_income", "operating_cash_flow"],
start="2015",
asof="2024-01-01" # Only data available as of this date
)
print(df.tail()) report_date revenue net_income op_cash_flow
0 2023-09-30 383,285,000,000 96,995,000,000 110,543,000,000
1 2022-09-24 394,328,000,000 99,803,000,000 122,151,000,000
2 2021-09-25 365,817,000,000 94,680,000,000 104,038,000,000
3 2020-09-26 274,515,000,000 57,411,000,000 80,674,000,000
4 2019-09-28 260,174,000,000 55,256,000,000 69,391,000,000
# All values sourced directly from SEC 10-K filings via EDGAR
All data is filtered by filing_acceptance_dt —
the exact timestamp the SEC received the filing. This eliminates look-ahead bias that corrupts most financial backtests.
Your models reflect what was actually known at each point in time, not what was known later.
Built directly from SEC filings. Fully traceable to the source document.
Traditional financial data platforms were built for institutions, not developers. The result is an industry that's expensive, opaque, and hostile to independent builders.
Bloomberg, FactSet, and Refinitiv charge tens of thousands per year — pricing out solo researchers, indie developers, and early-stage startups entirely.
Low-cost alternatives often have inconsistent data, undocumented normalization choices, and no clear sourcing — making it impossible to trust the numbers.
When a number looks wrong, you can't trace it. Most vendors offer no way to verify data against the original SEC filing that it came from.
Most datasets are not truly point-in-time. They silently include restated or revised data, corrupting backtest results with information that wasn't available at decision time.
By building directly from SEC EDGAR filings with full transparency, OpenFinDB provides datasets where every number traces to its source document, every timestamp reflects actual filing acceptance time, and the entire pipeline is open source and reproducible.
Every design decision optimized for correctness, traceability, and developer productivity.
Financial values are tied to the exact SEC filing acceptance timestamp, preventing look-ahead bias in research and backtesting. Know exactly what data was available at any moment in history.
filing_acceptance_dt
Every financial number traces back to the original SEC filing — including accession number, CIK, and XBRL concept tag. No black boxes. Complete transparency from raw filing to final dataset.
accession_number
Query financial data using simple Python APIs or SQL designed for quant research and fintech development. DuckDB-compatible Parquet files for fast offline analysis.
fundamentals.get()
OpenFinDB builds financial datasets directly from SEC filings using a transparent, reproducible pipeline. Each layer adds structure without losing traceability to the source.
Every layer is fully traceable. A signal in QuantDB can always be traced back through FundamentalsDB → EdgarDB → the original SEC EDGAR filing.
OpenFinDB is designed as transparent financial data infrastructure — every component is documented, reproducible, and built on proven open-source tools.
Point-in-time datasets using SEC acceptance timestamps — not report dates
Full XBRL traceability to concepts, filings, and accession numbers
Reproducible pipelines built from raw filings with deterministic processing
Developer-first APIs and Python libraries designed for quant workflows
Designed for builders and researchers who need financial data that is correct, transparent, and accessible.
Build factor models and backtests with point-in-time fundamentals. Eliminate look-ahead bias at the data layer, not as an afterthought.
Power financial applications with transparent, traceable datasets. Know exactly where every number comes from — and verify it independently.
Access reproducible financial data linked directly to SEC filings. Full methodology transparency for peer review and replication.
OpenFinDB includes open-source components that let the community inspect, extend, and contribute to the financial data pipeline. Transparency isn't just a feature — it's the architecture.
View EdgarDB on GitHubEDGAR ingestion system — archives raw SEC filings with full metadata and submission history.
Parsing framework for extracting structured financial data from XBRL-tagged SEC filings.
Documented schemas for FundamentalsDB and QuantDB, enabling community contributions and integrations.
OpenFinDB is being built incrementally — one transparent layer at a time.
Complete EDGAR ingestion pipeline. Archive all SEC filings with metadata, acceptance timestamps, and structured submission data. Foundation for all subsequent layers.
Normalized financial dataset extraction from SEC filings. Income statement, balance sheet, and cash flow data with full XBRL concept mapping and point-in-time accuracy.
Research-ready datasets and derived signals. Factor datasets, anomaly signals, and Parquet files for offline analysis. Python and SQL APIs for quant research workflows.
OpenFinDB is being built in the open. Star the repository to follow along as the data pipeline progresses.