O
OmniFocus
A

A private knowledge engine for complex research

Index thousands of PDFs, trace citations with semantic precision, and extract structured claims without sending proprietary documents to public clouds.

2,847researchers waiting for beta access

Join Cohort 04 of our Private Beta

We onboard 50 research teams every Monday. Access includes full API keys, local embedding models, and private sandbox tooling.

Launch Roadmap

Release & Deployment Roadmap

Track our engineering milestones as we progress toward public availability.

Private Engine Core (Alpha)

CompletedQ1 2026

Deterministic citation graph verification, local vector embedding engine, and isolated memory container architecture.

Cohort Beta & API Preview

In ProgressQ2 2026

Onboarding 50 teams weekly for synthetic extraction testing, CLI tooling, and SDK preview distribution.

Global General Availability

UpcomingQ3 2026

Desktop apps for macOS/Linux, enterprise on-prem clustering, and production SOC-2 certified sandbox guarantees.

Engineered for reproducible discovery

A preview of the architecture powering our upcoming release.

Core Capability

Deterministic Verification Chains

Every synthesized summary links directly to grounded paragraph spans. No phantom references or unverifiable claims.

Sub-50ms In-Memory Vector Search

Vector indexes run on your local NVMe storage. Query across 50,000 pages of literature with zero network roundtrips.

Core Capability

Zero Data Retention Sandbox

Workspaces operate inside ephemeral memory isolation. Documents and embeddings are discarded immediately upon session termination.

OmniFocus — Private Research & Knowledge Engine