01
Desktop app
A native desktop app with its own window. Download, sign in, and the agent is there.
Download the app →The open-source agent framework that evolves.
VAF is a local-first agent framework and harness in one — with persistent memory, a full toolset, and a living presence. It learns new skills as it works, and runs on your desktop, your own server, or wherever you build.
Open source · macOS, Windows & Linux
One agent, many capabilities Memory Web search Code execution Browser Files Self-learning Multi-agent
One framework, your way
Run it however suits you.
One agent, local-first by design. Run it on your machine, deploy it to your own server, or skip the infrastructure entirely and run it on the Veyllo API.
02
Self-hosted
Install VAF as a host on your own server. Your data, your keys, your infrastructure — fully under your control.
Read the docs →03
Build on the framework
VAF is a framework and harness in one. Extend it, wire in your own tools over MCP, and ship your own agents.
Explore the framework →No providers to wire up and nothing to host — the Veyllo API gives VAF every model and connection it needs behind a single key.
See it work
One agent. Every kind of work.
Chat, code, browse, research, organize files, draft documents — one agent that opens the right window for each task and does the work. Everything here is a live window, not a screenshot.
Why Veyllo
An agent that remembers, acts, and improves.
The capabilities that set the agent apart.
Persistent memory
Vector-backed memory lets the agent recall people, projects, and preferences across sessions — context that carries over instead of resetting on every restart.
A complete toolset
Search the web, run code, browse, and read and write files. Native tools plus anything you connect over MCP, unified in a single tool system.
Learns your tools
Connect a new or custom tool and the agent learns to use it — a stronger model teaches it the tool once, so it handles it reliably on its own from then on.
Local-first and private
Runs on your machine, in its own desktop window. Your data and keys stay where you put them.
Multi-agent by default
Delegates to sub-agents for parallel work and brings the results back together — multi-agent orchestration, built in.
A face with state
The living dot isn’t decoration. It signals exactly what the agent is doing — thinking, searching, waiting, blocked, done.
Veyllo API · preview
Every model VAF needs, behind one key.
VAF runs on LLMs and the services it plugs into. Normally you’d sign up with each provider and wire in their API keys yourself — or host the models locally on hardware you may not have. The Veyllo API replaces all of that with a single key.
✓
One key instead of many. Every model and provider VAF talks to, behind a single API — no scattered accounts or keys to manage.
✓
Nothing to host locally. No GPUs, no self-run model servers — compute and model hosting run on Veyllo.
✓
Built for production. Keys, rate limits and usage metering from day one.
Open development
Follow the build on GitHub.
Read the code, track the roadmap, and file issues. Development happens in the open.
Active
development
Issues
open & tracked
Roadmap
public
Put the agent to work.
Download the desktop app, or build on the Veyllo API.