Epstein Files Transparency Act
Public Resources & Official Records
Survivor Protection
Protecting survivors who have not voluntarily identified themselves is our highest priority. The DOJ released nearly one million documents with inconsistent and often inadequate redactions — survivor names, photos, passport details, and contact information that should have been protected were left exposed throughout the corpus. Despite the automated guardrails and manual curation we have put in place, this site processes those documents using facial recognition and generative AI, which means we may inadvertently surface survivor identities that the DOJ failed to redact.
If you believe we are exposing a survivor’s identity, please email us immediately
so we can take corrective action within 24 hours:
To help us act quickly, please include as much of the following as possible:
- The EFTA ID or page URL where the exposure appears
- The type of content to remove (name, photo, passport, contact details, email, etc.)
- The page number within the document, if applicable
We will never ask you for private information or proof of identity to process a removal request. You do not need to identify yourself as a survivor — third-party reports on behalf of survivors are equally welcome. See our full Survivor Protection & Takedown Policy.
EFTA Document Lookup 🔗
Enter an EFTA ID to get the direct DOJ download link:
Official Records & Analysis 🔗
Document Findings
Criminal evidence against Leon Black and Jes Staley, victim diary entries (forensically verified), island visit documentation, prosecution memos, financial records showing $100M+ Wexner embezzlement, and more.
New EvidenceDiary Transcriptions
Handwritten victim diaries with full transcriptions. Includes forced pregnancy documentation, coded messages with rail fence cipher decoding, and interview notes naming Donald Trump, Harvey Weinstein, and Prince Andrew.
NewFinancial Network
Interactive graph of Epstein's financial infrastructure — $158M+ Leon Black payments, $177M Deutsche Bank flows, $1B+ JPMorgan suspicious activity, trust structures, and shell entities.
TrackingReal-World Outcomes
Resignations, suspensions, investigations, and other consequences triggered by the EFTA releases.
InteractivePeople
400+ profiles of individuals connected to Epstein, with document counts, key findings, network connections, and AI-generated summaries sourced from the EFTA corpus.
InteractiveTimeline
Chronological timeline of key events from the Epstein case, arrests, convictions, and document releases.
InteractiveFlights
Flight records from Epstein's aircraft — helicopters and jets — extracted from EFTA documents.
InteractiveLocations
Interactive map of properties connected to Jeffrey Epstein — NYC townhouse, Palm Beach, Little St. James, Zorro Ranch, and more.
TrackingDeleted Files
Files that have been removed from the DOJ website after publication, including a photo of Donald Trump.
Local ArchiveVideo Gallery
Video files from the DOJ disclosures, with metadata filtering.
🔊 Note: Video audio has been transcribed using OpenAI's Whisper large-v3 model.
Local ArchiveImage Gallery
665,000+ images extracted from PDFs. Filter by dataset, faces, redactions, or person. Facial recognition applied to identify individuals.
Local ArchiveAudio Gallery
Audio files from the DOJ disclosures, with transcriptions.
🔊 Note: Audio has been transcribed using OpenAI's Whisper large-v3 model.
Local ArchiveDocuments Gallery
Spreadsheets and other non-PDF documents from the disclosures.
Local ArchiveThumbs.db Gallery
83 thumbnail images extracted from two Windows Thumbs.db database files (EFTA01242503, EFTA01242504) in Dataset 9.
TranscriptLes Wexner Deposition — Epstein Probe
Full transcript of Les Wexner's deposition on the Epstein probe (4h 56m). Transcribed with Whisper large-v3. Watch on YouTube →
TranscriptsHouse Oversight — Media Transcripts
81 DOJ-OGR audio and video files from the House Oversight Committee release, transcribed with Whisper large-v3. Includes videos from NATIVE006, NATIVE008, NATIVE011, and NATIVE012.
FAQFrequently Asked Questions
Common questions about the black book, email attachments, document numbering, and navigating the disclosures.
Dataset Overview 🔗
| Set | Files | Pages | Images | Videos | Audio | Deleted | Metadata | Archive | Errors |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 3,158 | 3,158 | ✓ 3,158 | 0 | 0 | 9 | 3,158 | 3,144 | 0 |
| 2 | 575 | 699 | ✓ 698 | 1 | 0 | 8 | 575 | 577 | 1 |
| 3 | 96 | 1,847 | ✓ 1,717 | 0 | 0 | 1 | 96 | 69 | 0 |
| 4 | 179 | 2,704 | ✓ 2,249 | 0 | 0 | 0 | 179 | 154 | 0 |
| 5 | 141 | 120 | ✓ 120 | 0 | 0 | 1 | 141 | 122 | 0 |
| 6 | 19 | 487 | ✓ 484 | 0 | 0 | 1 | 19 | 20 | 0 |
| 7 | 26 | 660 | ✓ 660 | 0 | 0 | 0 | 26 | 19 | 0 |
| 8 | 11,055 | 29,348 | ✓ 27,189 | 419 | 4 | 2 | 11,055 | 11,036 | 362 |
| 9 | 539,237 | 1,223,774 | ✓ 272,201 | 4,068 | 216 | 1,879 | 533,606 | 531,417 | 1,983 |
| 10 | 504,141 | 950,106 | ✓ 288,024 | 829 | 44 | 0 | 504,032 | 504,539 | 874 |
| 11 | 336,769 | 517,416 | ✓ 67,737 | 4 | 0 | 293 | 336,769 | 331,661 | 4 |
| 12 | 2,520 | 1,596 | ✓ 815 | 0 | 0 | 4 | 2,520 | 154 | 0 |
| Total | 1,397,916 | 2,731,915 | ✓ 665,052 | 5,321 | 264 | 2,198 | 1,392,176 | 1,382,912 | 3,224 |
Note: Page counts are for PDF files only. Image counts are images extracted from PDFs.
Files = URLs listed on DOJ site. Metadata = unique files confirmed by scraper (deduplicated). Archive = files preserved in torrent ZIP archives. Errors = files that failed metadata extraction (corrupt or non-PDF). DOJ deletions and modifications tracked separately.
The DOJ deleted EFTA00000468, which featured a photo of Donald Trump. That is why we track deleted files. View deleted files
EFTA ID Ranges 🔗
EFTA IDs range from 1 to 2.7M. Black indicates gaps in the disclosed IDs.
Download efta_page_stamps.json (80 MB) — maps 369K parent documents to their 1.35M child pages across the corpus.
PDF Creation Date Analysis 🔗
PDF metadata records when a document was digitised. This reveals how the DOJ processed the corpus before releasing it.
| Date | Files | Notes |
|---|---|---|
| No date | 554,991 | Pre-existing native digital files (emails, spreadsheets, born-digital PDFs) — no scan date embedded |
| Apr 2013 – Jan 2014 | 148 | Original early-digitisation documents; oldest files in the corpus |
| December 2025 | 10,173 | DOJ early-stage scanning ~6 weeks before the statutory release deadline |
| Jan 1–14, 2026 | 50 | Smaller processing runs in early January |
| January 29, 2026 | 94,630 | 656,914 files scanned in just two days, immediately before the January 31 public release — the bulk of the corpus digitised in a last-minute sprint |
| January 30, 2026 | 562,284 | |
| Jan 31 – Feb 1, 2026 | 612 | Final cleanup and additions processed after the release date |
| Total with date | 667,897 | All processed by DOJ’s OmniPage CSDK 21.1 OCR pipeline |
Creation dates are from the CreationDate field in PDF metadata. The remaining ~134K files are those for which metadata could not be retrieved (encrypted, corrupted, or download failures).
Oversight & International Response 🔗
DOJ Withheld & Removed Epstein Files Related to Trump
An NPR investigation and New York Times review found the DOJ withheld over 50 pages of FBI interviews with a woman who accused Trump of sexual assault as a minor in the 1980s. Of four FBI interviews conducted, only one was released — and it omits the Trump accusation. Documents mentioning a second accuser were temporarily removed from the database. Rep. Robert Garcia confirmed the files are also missing from unredacted versions reviewed at DOJ. House oversight Democrats have launched an investigation into whether the DOJ purposely withheld these materials. First reported by journalist Roger Sollenberger.
UN Experts: Epstein Files Suggest Global Criminal Enterprise
Ten UN Special Rapporteurs declared the files contain “disturbing and credible evidence” of systematic abuse that may meet the legal threshold of crimes against humanity. They demand prosecution in all competent national and international courts and warn that resignations alone are not adequate accountability.
SurveillanceDOJ Tracked Lawmakers' Epstein File Searches
AG Bondi was photographed holding a printout of Rep. Jayapal's search history from the DOJ reading room. Bipartisan outrage followed, with lawmakers demanding the DOJ cease monitoring which members review which files.
ContemptKhanna & Massie Threaten Inherent Contempt Against Bondi
Bipartisan resolution drafted to hold the Attorney General in "inherent contempt" for failing to release all eligible Epstein files by the statutory deadline, proposing daily fines until compliance.
CongressSpecial Master Request to Compel Full Release
Reps. Khanna and Massie called for appointment of a special master to oversee the DOJ's release, alleging the agency violated transparency requirements through missed deadlines and excessive redactions.
FOIA LawsuitDemocracy Forward Sues DOJ Over Epstein Files
First-of-its-kind lawsuit challenging the Trump administration's handling of Epstein files through FOIA, seeking records the government has not released expeditiously.
CourtSecond Circuit Remands Epstein FOIA Case
Federal appeals court signaled remand of a FOIA case seeking FBI Epstein investigation documents, ordering reconsideration of disclosure obligations under the new Transparency Act.
PBSRaskin Accuses DOJ of “Spying” on Lawmakers
Rep. Jamie Raskin accused the Justice Department of surveilling members of Congress reviewing Epstein files at DOJ-controlled terminals, calling it an attempt to obstruct and intimidate.
External Resources 🔗
DOJ Epstein Library
The Department of Justice's official repository for released documents, court records, and FOIA disclosures.
Official SourceDOJ Press Release: 3.5M Pages Published
Official announcement of the Department of Justice's publication of 3.5 million responsive pages in compliance with the Epstein Files Transparency Act.
OverviewWikipedia: Epstein Files Transparency Act
A comprehensive overview of the legislation, its history, passage, and implementation details.
OverviewWikipedia: Epstein Files
Overview of the released documents, key findings, notable individuals named, and public response.
Legislative TextCongress.gov: H.R. 4405 Text
The full official text of the Epstein Files Transparency Act as enrolled and signed into law.
Congressional ReportEFTA Section 3 Report to Congress
Official report including the list of all government officials and politically exposed persons identified in the EFTA releases.
Legal Archives & Court Records 🔗
FBI Vault: Jeffrey Epstein
22 parts of FBI investigative files released under FOIA. A separate release channel from the EFTA documents.
FOIAThe Black Vault: FBI & Unsealed Docs
FBI records on Jeffrey Epstein obtained through FOIA, plus Ghislaine Maxwell and Virginia Giuffre unsealed documents.
Court RecordsThe Black Vault: 2019 Unsealed Docs
2,024 pages of court documents from Giuffre v. Maxwell unsealed Aug 2019. Includes a searchable processed PDF (394 MB).
Primary SourceVirginia Roberts Giuffre: Memoir
72-page unpublished memoir by Virginia Roberts Giuffre, unsealed from the Giuffre v. Maxwell defamation case in 2019.
FOIAMuckRock: Palm Beach PD Records
Palm Beach Police Department investigation files from the original 2005 Epstein case. 11 files including probable cause affidavits.
Court RecordsDocumentCloud: Unsealed Docs
Court documents from Giuffre v. Maxwell unsealed in January 2024. Depositions, flight logs, and witness testimony.
Court RecordsGiuffre v. Maxwell Docket
Browsable docket: 1,360 entries, 2,991 documents from the civil case that produced the unsealed documents. Sourced from CourtListener / RECAP.
FOIAMuckRock: FOIA Requests
Public FOIA requests related to Epstein filed with US Marshals Service, Palm Beach PD, Secret Service, and other agencies.
GovernmentHouse Oversight Committee
Official release page for Epstein records provided by the DOJ to the House Oversight Committee, including estate documents and related files.
Leaked DataEpstein Email Archive (DDoSecrets)
~20,900 emails from Epstein's personal Yahoo account (2002–2019), independently obtained by the nonprofit transparency collective Distributed Denial of Secrets. Unredacted originals of emails that appear in the EFTA corpus as redacted PDF printouts.
Research Tools 🔗
DOJ Files on Pinpoint
Search the Department of Justice Epstein Files releases. This includes the more recent releases from the DOJ investigation.
Search ToolEstate Files on Pinpoint
Search the Epstein Estate Documents released by the House Oversight Committee. Contains roughly 20,000 pages of emails, schedules, and other records.
ArchiveEpstein Files on GitHub
Community archive with torrent files and mirrors of the DOJ releases for offline access and preservation. Note: Some torrents are incomplete.
ArchiveEpstein Files Change Tracker
Tracks DOJ modifications and deletions across all 12 datasets by archiving HTTP metadata (ETags, Last-Modified headers). Documents which files were altered or removed after publication.
ExternalAdvanced Network
A more comprehensive and interactive network visualization with additional features.
AnalyticsDOJ Website Analytics
Real-time analytics showing the most accessed files and pages on justice.gov, including Epstein documents.
AnalyticsSite Analytics
Public analytics dashboard for this site — visitor traffic, popular pages, and usage trends.
Open SourceEpstein Archive
Open-source archive with AI-generated summaries and entity extraction for 8,175 processed documents. MIT licensed.
Search ToolEpsteinWeb
Search Epstein emails and trace documented connections with degrees-of-separation mapping across FOIA releases and court filings.
Open SourceEpstein Search
Open-source full-text search tool for the Epstein files. Search across all 12 DOJ datasets with instant results.
CommunityEpstein Files Research Discord
Community server for collaborative research, document analysis, and discussion of the EFTA releases.
CommunityCorroborators Wiki
Non-political collaborative wiki to surface details, connections, and data from officially released Epstein files. Includes per-document pages, people categories, and a public forum.
Open SourceThe PDF Files
Open-source indexing of 900K+ DOJ documents with full-text search, relationship mapping across 148+ individuals, and a 2,500+ event timeline. Public domain.
Open SourceEpstein Research (Data Science)
100+ forensic analysis reports covering financials, named individuals, and evidence analysis. All claims cite specific EFTA documents. AI-assisted, not independently verified.
ExternalEpstein Visualizer
Interactive network visualization of connections between individuals in the Epstein documents.
ArchiveEpstein Files Explorer
Browsable archive of all 12 DOJ datasets with direct links to original government sources. Organized by release date and data set.
TechnicalPDF Forensics Case Study
PDF Association technical analysis of the Epstein PDF documents, examining metadata, redaction methods, and document structure.
Media Coverage 🔗
Miami Herald
Julie K. Brown's "Perversion of Justice" investigation that broke the Epstein story open in 2018 and led to his arrest.
NewsThe New York Times
Breaking news, investigations, and archival coverage regarding Jeffrey Epstein and associated cases.
NewsThe Guardian
Ongoing reporting and special reports on the Epstein inquiry and global impact.
Data Downloads 🔗
Raw metadata, URL lists, and OCR-extracted text for researchers and developers.
Metadata (JSONL) & URL Lists (TXT)
Media Metadata
Video and audio metadata extracted with ffprobe. Includes codec, resolution, duration, bitrate, creation dates, device model, GPS coordinates, and transcripts.
PDF Text Extraction (OCR)
OCR-extracted text from PDFs using pdftotext. Some files are missing due to DOJ deletions.
Acknowledgments 🔗
Maria Farmer Drawing Identifications
Thank you to CosmicGoddess777 and BubsyBot on Reddit for identifying the people depicted in Maria Farmer's drawings from the Epstein documents (EFTA01652467, EFTA01652669).
See also: Everyone I Could Identify · The Setiles Pt. 2
Face & Photo Identifications
Thank you to Broka for lightning-fast help identifying faces in the Epstein documents, and to doodleDee and Vii from the Epstein Files Research Discord for help with photo identifications.
Data Processing
Thank you to Korroni for help with deduplication of JSONL files, EFTA page-stamp mapping, and much more.