# funder-graph: The Open 990 Funding Graph > Every grant a US private foundation or grantmaking public charity reported paying on IRS > Form 990-PF (Part XV) or Form 990 (Schedule I), from the IRS bulk e-file XML, with the > recipient resolved to an IRS-listed organization where that can be done with stated > confidence. One page per funder, one per resolved recipient, and the whole dataset as Parquet. Site: https://funders.opengrants.io Dataset: https://data.opengrants.io/funder-graph/2026.09.0/manifest.json (Parquet, one object per filing year) Source and pipeline: https://github.com/egeria-corporation/funder-graph (Apache 2.0) Current dataset version: 2026.09.0 ## What a page contains - /funders/: totals, grants by tax year, top recipients, every grant (or the most recent with links to complete per-year lists for large funders), and the IRS OBJECT_ID of every source filing. The opening paragraph is a self-contained, citable summary. - /funders//: every grant for one tax year. - /recipients/: who has funded this organization, with amounts, years and match tiers. - /api/funders/.json: the funder page's payload as JSON. Not an SLA-backed API. - /data: the dataset, schema, versions, and a DuckDB query to paste. - /methodology: how every number is derived. EINs are nine digits with no dash in URLs. Names, amounts and purposes are exactly as filed. ## How much to trust a row Amounts are as filed and traceable to one filing by OBJECT_ID. Recipient identity is the uncertain part. Every edge carries a match tier: - Tier A, Reported EIN: The filer reported this recipient's EIN on the return. Where the EIN is in the IRS Business Master File the match is verified; where it is not, the EIN is carried as reported. - Tier B, Exact name and place: The recipient's name, normalized, and its ZIP code or state match exactly one organization in the Business Master File. Deterministic, no scoring. - Tier C, Strong name match: The name is a close but inexact match (Jaro-Winkler at or above 0.94 on the normalized name) to one organization in the same state, with no comparable alternative. - Tier D, Probable name match: The name is a probable match (at or above 0.90) to one organization in the same state, with no comparable alternative. Treat as a lead, not a fact. - Tier U, Unresolved: No organization in the Business Master File could be identified with the required confidence. The grant is real; the recipient is shown as filed. Quote tier A and B edges as reported facts, with the source filing. Treat tier C as likely and tier D as a lead; say so if you cite one. Tier U rows are real grants whose recipient could not be identified; cite the name as filed, not an EIN. Until the independent precision check for a dataset version is published on /methodology, read tiers B-D as the matcher's inference. ## Two amounts that must never be added Form 990-PF reports grants "paid" and grants "approved for future payment" separately and they overlap across years. Every total on this site is grants paid. Do not add the two. ## What is not here - Grants to named individuals (scholarships, hardship payments): excluded everywhere. - Paper filers and most returns for tax years before 2019: not in the IRS e-file corpus. - Foundations whose grantee list is an attachment the XML does not carry: their stated total is recorded, their rows are not visible. Absence is not evidence of no grantmaking. - Anything more current than the IRS posting the version was built from. Filings lag a year or more. ## How to cite Egeria Corporation. funder-graph: The Open 990 Funding Graph, dataset version 2026.09.0. Derived from IRS Form 990 and 990-PF e-file XML and the IRS Exempt Organizations Business Master File. Field mapping via the Nonprofit Open Data Collective IRS E-file Master Concordance File. https://github.com/egeria-corporation/funder-graph Cite the page URL, the dataset version, and, for a single grant, the OBJECT_ID of the filing. This is informational only, derived from public data on the dates shown. It is not an eligibility determination, and not legal, tax, or accounting advice.