Does Feluda send my data to any server?

No. Feluda itself never sends your data anywhere. If you choose a cloud AI provider like OpenAI, your prompts go to that provider's API — but if you use a local model, everything stays on your machine. Feluda never collects, uploads, or transmits your workflow data to its own servers.

What local AI models can I use with Feluda?

Feluda supports Ollama and LM Studio out of the box. Both let you download and run open-source AI models locally — no API key, no cloud account, no data leaving your computer. On paid plans, you can also connect custom OpenAI-compatible endpoints for self-hosted models.

Run AI Workflows Without Sending Data to the Cloud

Q: How can I run AI workflows without sending data to the cloud?

Feluda is a desktop application that runs entirely on your computer. When you use local AI models like Ollama or LM Studio, your data never leaves your machine. You build workflows visually, run them locally, and your files, prompts, and results stay on your desktop.

Your data is yours. Feluda runs on your desktop — build, execute, and schedule AI workflows without a single byte leaving your machine.

Local models, encrypted secrets, visual pipelines — all on your computer.

The Cloud Problem

Every time you paste a document into ChatGPT, submit text to an AI API, or upload a file to a cloud automation service, your data travels to somebody else's server. You trust that they handle it responsibly, but you cannot verify it. For sensitive contracts, medical records, internal reports, or customer data, that trust is a risk you may not want to take.

Running AI workflows without the cloud means your data stays where it belongs — on your machine, under your control, visible only to you.

How Feluda Keeps Your Data Off the Cloud

🖥️

Desktop-Native Application

Feluda is not a web app. It installs on your Windows, macOS, or Linux machine and runs entirely on your desktop. Your workflows, files, and results live on your local file system — not on a remote server.

🤖

Local AI Models

Connect Feluda to Ollama or LM Studio — both run open-source AI models directly on your hardware. No API key. No internet connection needed. No data leaves your machine. Learn more about local AI in Feluda.

🔐

Encrypted Secrets Vault

API keys and credentials are encrypted in your operating system's secure vault (macOS Keychain, Windows Credential Manager, Linux Secret Service). They never leave your machine. AI models never see them — Feluda injects them at runtime.

🚫

No Auto-Sync, No Telemetry

Feluda never uploads your data automatically. Synchronisation with your feluda.ai account only happens when you manually press the Sync button — and even then, it syncs purchased Genes, not your workflow data or results.

Feluda AI Providers — configure local and cloud providers — AI Providers: choose between cloud providers and local models — your choice determines where data goes

You Choose Where Data Goes

Feluda is provider-agnostic. The same workflow can run entirely locally or use a cloud API — the choice is yours, per block, per workflow:

🏠

Fully Local

Use Ollama or LM Studio for every AI block. Zero data leaves your machine. Zero API costs. Works without internet.

☁️

Selective Cloud

Use a local model for sensitive steps (PII extraction, classification) and a cloud provider for non-sensitive steps (image generation, summarisation). Mix privacy with capability.

🏢

Self-Hosted

On paid plans, connect to your own OpenAI-compatible endpoint — a model running on your team's server. Data stays within your infrastructure.

Frequently Asked Questions

How can I run AI workflows without sending data to the cloud?

Install Feluda on your desktop. Connect it to a local AI model like Ollama or LM Studio. Build your workflow in Studio and run it. Your data stays on your machine — nothing is sent to any cloud service.

Does Feluda send any data to its own servers?

No. Feluda never collects, uploads, or transmits your workflow data, results, or files. The only network activity is when you choose to use a cloud AI provider's API, or when you manually sync purchased Genes.

Can I mix local and cloud models in one workflow?

Yes. Each block in a workflow can use a different provider. Use a local model for privacy-sensitive steps and a cloud model for tasks that benefit from larger models — all in the same pipeline.