Sovereign AI Workspace
Your models. Your machine.
One sovereign workspace.
YveChat connects a hosted workspace to AI models running on your computer, while your models, conversations, images, and compute remain on your machine.
- Local models via Ollama
- Six-digit pairing
- Data stays on your machine
Local models, served by Ollama
Own the word. Sovereignty.
Money proved the thesis. Intelligence is the second application.
A policy about what a host chooses to do with your data.
A physical fact about where data sits and who can reach it.
YveChat is built on custody.
Own your compute
Ollama runs open models on your hardware. Prompts, output, and weights stay on your machine.
Own your data
Conversations, generated images, and credentials persist on your computer, in your custody.
Own your access
A six-digit code pairs the workspace to your machine, and you approve it locally.
YveChat is the sovereign AI workspace for people who run or control their own AI infrastructure. It suits local-AI users, self-hosters, builders, and privacy-conscious operators who are ready to install a local component.
A short checklist.
YveChat runs local models through a component on your computer. Five things get you started.
- 01
A supported desktop computer
Windows, macOS, or Linux with enough memory for the models you choose.
- 02
YveChat Local Bridge
The connector that links the hosted workspace to your machine. Public installers are on the way, so the preview uses a local build today.
- 03
Ollama or another configured runtime
Ollama is the primary supported runtime. vLLM and OpenAI-compatible endpoints have experimental support.
- 04
At least one downloaded model
Pull a model with Ollama before you begin. Availability depends on what you install.
- 05
A compatible browser with local-network permission
A current browser that can reach your local runtime once you grant permission.
Four layers. One machine.
Interface, bridge, runtime, and data. Hover a layer to isolate it, and switch the mode to recolour the page.
What leaves your machine
Nothing
Follow the data.
Four paths, stated plainly. Each row shows what leaves your machine.
Local models
AvailablePrompts, output, and model weights stay on your computer.
Nothing
Hosted providers
ExperimentalWhen you connect a hosted provider with your own key, that request reaches the provider you chose.
Prompts and output, to your provider
Web search
AvailableA search query travels through the local tool gateway to the search provider, and results return with sources.
Your query, to the search provider
Hosted interface
AvailableThe workspace is static files served from yvechat.xyz. Models and conversations stay on your machine.
Only the page load
Every capability, with its status.
Available runs today. Experimental works with limits. Planned sits on the roadmap.
Local language models
Ollama serves open models on your hardware. Prompts, output, and weights stay on your machine.
Local Bridge pairing
A six-digit code links the hosted workspace to your computer, and you approve it locally.
Local image generation
ComfyUI renders images on your machine through image workflows discovered on your system.
Web search with citations
Web search runs through the local tool gateway and returns clickable sources.
Local and wallet sign-in
Sign in with a local account. Wallet authentication is available and optional.
OpenAI-compatible and vLLM endpoints
Configured OpenAI-compatible and vLLM endpoints have limited support today.
Hosted-provider key management
Full management of hosted-provider API keys is in progress.
MCP connections
Model Context Protocol connections sit on the roadmap.
Document reading
Reading and retrieval over your documents sits on the roadmap.
Sandboxed code execution
Running code in a local sandbox sits on the roadmap.
Tools that run on your machine.
Web search runs today through the local tool gateway and returns clickable sources. Image generation renders locally through ComfyUI. Document reading and code execution sit on the roadmap.
Pair the workspace to your machine.
Local Bridge runs on your computer and connects the hosted workspace to your local runtimes through an explicit pairing flow. A six-digit code links the two, and you approve it locally.
Pairing flow
- 1Local Bridge runs on your computer.
- 2The workspace shows a six-digit code.
- 3You approve the code on your machine.
- 4The workspace reaches your local runtimes.
What stays on your machine
- Models
- Conversations
- Generated images
- Credentials
- Execution
Why a bridge exists
Ollama origins
Ollama rejects cross-origin requests by default. Reaching it from a hosted page needs OLLAMA_ORIGINS set and the daemon restarted.
Read the sourceChrome Local Network Access
Chrome gates access from public sites to loopback and private ranges behind a permission prompt, now extended to WebSocket and WebTransport.
Read the sourceWebKit loopback history
WebKit carried a loopback mixed-content issue since 2017. Safari 18 blocked HTTPS pages from loading http://127.0.0.1 until a March 2025 fix.
Read the sourcePublic Local Bridge installers for Windows, macOS, and Linux are on the way. The developer preview runs a local build today.
Four steps to your first local model.
Set up the Local Bridge, connect a runtime, pair the workspace, and pick a model.
Install YveChat Local Bridge
Run the Local Bridge on your computer so the workspace can reach your machine.
Install or connect Ollama and download a model
Set up Ollama and pull at least one model. Larger models need more memory and stronger hardware.
Open the app and approve the six-digit pairing code
Open the workspace, then confirm the six-digit code on your machine to pair.
Select a detected model and begin
YveChat discovers your installed Ollama models after the Bridge connects. Pick one and start.
Local Bridge required
When something stalls.
Six common states and the step that clears each.
Local Bridge unavailable
Confirm the Bridge is running on your computer, then pair again. Restart it if the workspace cannot reach it.
Browser local-network permission denied
Grant local-network access in the browser prompt, then reload the workspace.
Ollama unavailable
Start Ollama and confirm it is listening. The Bridge reaches it once it is running.
No models detected
Download a model with Ollama, then refresh. YveChat discovers installed models after the Bridge connects.
Pairing code expired
Request a new six-digit code and approve it on your machine.
Model too large for available hardware
Choose a smaller model. Larger models need more memory and stronger hardware.
Hosted interface, local machine.
Four layers, each with its state. The hosted page pairs to a bridge on your computer, and your runtimes and data stay local.
// hosted page, paired to your machine
Hosted-provider key management is experimental. MCP, document reading, and sandboxed code execution are planned.
Hosted Interface
AvailableA static workspace served from yvechat.xyz. Models and conversations stay on your machine.
> interface = static.hosted
Local Bridge
AvailableA connector on your computer, paired by a six-digit code with your local approval.
> bridge.pair = six.digit.code
Local Runtimes
AvailableOllama serves language models and ComfyUI renders images, both on your hardware.
> runtime = ollama + comfyui
Local Data
AvailableConversations, images, credentials, and execution stay on your machine.
> data.location = your.machine
Self-custody, now for your intelligence.
The principle that reshaped money applies to thought.
When the model runs on your machine, the workspace is yours to command.
YveChat carries self-custody into the way you work with AI.
Five eras to full sovereignty.
Each era hands another layer of control to the user.
Custody
Local models, local data, and local pairing on your own machine.
BridgeNow
Local inference through a signed, explicitly paired connection. Public installers are next.
Gateway
Web search runs today. Document reading and sandboxed code execution come next.
Network
An opt-in compute network for shared capacity.
Sovereignty
Governance over protocol parameters and independent oversight.
Utility across the five eras.
One utility arrives per era, each tied to a capability the network gains. Every $YVE utility is planned.
$YVE has not been generated. No contract is deployed and no sale is open. Any offer of $YVE is fraudulent.
Sovereign Pass
Holding or staking unlocks the scaling tier: extended agents, team workspaces, encrypted sync, priority builds, and deployment support.
Gateway Credits
Meters tool calls routed through YveChat-operated providers. Your own infrastructure is exempt.
Marketplace Settlement
Settles paid listings and revenue share for agents, prompt packs, and configurations.
Compute Settlement
Pays operators serving models to others through the opt-in compute network.
Governance
Weight over protocol parameters, emissions, treasury allocation, and marketplace policy.
Proof you can run.
Four claims you can check for yourself.
Local models answer with nothing leaving your machine.
Open your network inspector while a local model answers. Watch for zero outbound requests.
The workspace pairs only after you approve a code.
Start the pairing flow. The workspace connects once you approve the six-digit code on your machine.
A local model answers offline.
Disconnect from the network and run an Ollama-served model. It answers.
Model availability matches what you installed.
Pull a model with Ollama, then reload. YveChat lists the models present on your machine.
Straight answers.
Your models.Your machine.Your workspace.
Follow the setup to run local models today. Read the docs for the full architecture.
Built for people ready to run their own AI.