Local-first desktop app · macOS · Windows & Linux alpha
Korvo is a private research workspace that runs on your machine. Drop in decks, PDFs, notes and transcripts — it organises the context, plans the analysis, cites material claims, and keeps a record of how you decided.
No account needed — the download starts right away.
🔒 Your workspace stays on your machine · Explorer free · Pro $19/mo or $149/yr
Explorer
Free
Local workspace, core analysis
Pro
$19/month
or $149/year — save $79 a year. Renews at the same rate while the subscription stays active; planned public price $299/year or $29/month.
Everything in Explorer, plus
Korvo does not resell model tokens. Cloud-provider API usage is billed separately by the provider you choose.
One project, end to end
Capture → Plan → Verify → Challenge → Decide
Scroll to follow the project →
12 files: pitch deck, financial model, founder notes, market research, and two call transcripts.
Files
Project library
12 sources · indexedKorvo detects the goal, makes assumptions visible, flags missing context, and proposes the work.
Detected goal
Evaluate Series B opportunity — unit economics, retention, comps
Missing: cohort churn · cap table
Approve planEvery material claim links back to the evidence selected from the project.
Burn multiple is 2.4× on trailing twelve months.
grounded“Net burn: $1.2m · Net new ARR: $500k · TTM”
Open source in workspace → jump to selected span
One model builds the case. A second critiques it. The revision and unresolved questions stay visible.
Model A
Efficient growth and 2.4× burn support the round at the ask.
Model B
Retention evidence is cohort-incomplete; concentration risk underweighted.
Merged
Conditional proceed — pending churn diligence and top-10 customer check.
The conclusion, evidence, caveats, model details, and timestamp remain together for later review.
Reason: strong unit economics, subject to validating gross retention and customer concentration.
Coming next
Korvo agents work across the tools you approve and deliver results to your private workspace.
Starting with Google Calendar, Drive, and Korvo Workspace.
Start every day with a briefing prepared from your calendar, approved Drive folders, and relevant Korvo projects.
Every weekday at 8:00 AM
Permissioned
Access is scoped to the accounts and folders you approve.
Private cloud execution
Each run happens in an isolated environment.
Visible run history
Every run, source, and result stays inspectable.
Be among the first to test Korvo Agents and help shape the product.
What Korvo is
Chat gives you an answer. Korvo helps you define the job before the answer is written.
A workspace stores your material. Korvo connects each conclusion back to the material that supports it.
A decision engine keeps the plan, challenge, evidence, and final call together so another person can inspect them later.
The workflow
Drop files, notes, and pasted text into one project. Keep the messy source material together.
Korvo indexes the project locally so you can search and retrieve the right context.
It identifies the goal, assumptions, missing context, and analysis steps. You approve the plan before execution.
Produce briefs and syntheses grounded in evidence selected from your workspace, with citations back to the source.
Keep the conclusion, source links, model details, revisions, and unresolved questions together.
Built for
Run diligence across decks, financials, founder notes, calls, and market research. Pressure-test the case before it reaches partner review.
Turn board material, customer evidence, and scattered strategy notes into a plan with explicit assumptions.
Synthesize reports and project context into an actionable brief without losing the evidence behind it.
Organize a literature set, compare claims, and produce cited synthesis from the source material.
Privacy and models
Your workspace, files, local index, and decision history stay on your machine. When using a cloud model, selected context goes directly to your provider under your own key and agreement. Or connect Ollama and nothing leaves your machine at all.
Files, the local SQLite index, notes, plans, and decision history reside strictly on your device. Korvo never holds your workspace data on its servers.
Connect OpenAI, Anthropic, Google Gemini, Moonshot (Kimi), or OpenRouter directly with your own key. Prompts and selected context go to that provider under their terms. Korvo never proxies, inspects, or logs them.
Connect Ollama for full offline operation on local models. No prompt, project context, or document data leaves your machine.
Export your work to Markdown or JSON on any plan. Delete local workspace data anytime. Your files, notes, and generated outputs remain yours.
Korvo Embed
NewKorvo Embed is our on-device embedding engine. It understands your documents by meaning rather than keywords — a 34 MB encoder that downloads once, runs on your machine, and is structurally incapable of sending a document anywhere. It ships in the Korvo app today, we license it for other products, and it is coming to the Korvo API Platform.
Ask in your own words and find the document that says the same thing differently.
No HTTP client in the dependency tree — verifiable in one command, enforced in CI.
No account, no API key, no rate limit. Your index is a file on your disk.
Medha
Medha is Korvo’s on-device quality-measurement layer. It tests retrieval with deterministic metrics such as recall, MRR, and nDCG without model calls or private corpus content leaving the device. It does not generate answers or claim that an answer is true.