Install Ollama and download the starter model
Install Ollama using the native download, launch the app, and leave it running. In a new PowerShell window, pull the exact local model below.
Install Ollama Starter modelollama pull qwen3.5:4bCompute / WORKFLOW
Set up a local AI server with Open WebUI
Run a persistent, password-protected chat workspace on your own computer. Connect Open WebUI to local Ollama, start with Qwen3.5 4B, and confirm your conversation survives a restart.
EstimatedSource checked 9/29/2026
LOCAL
Before you begin
Hardware
Memory for one person using Open WebUI plus the selected Qwen3.5 4B starter. The interface and the model are separate processes.
| Machine | Open WebUI + Qwen3.5 4B |
|---|---|
| Windows PC · 16 GB RAM | Runs locally; a supported 8 GB GPU improves model responsiveness |
| Linux PC · 16 GB RAM | Runs locally; NVIDIA / supported AMD / Intel via Ollama |
| Apple Silicon · 16 GB unified memory | Suitable starting point; Ollama can use Metal |
| CPU only · 16 GB RAM | Works for local chat; replies take longer without GPU acceleration |
These are estimates for the bundled starter model, not a capacity guarantee for other models or multiple simultaneous users.
LOCAL AI SERVERSOURCE-LINKED SETUP
Set up a local AI server with Open WebUI
Run a persistent, password-protected chat workspace on your own computer. Connect Open WebUI to local Ollama, start with Qwen3.5 4B, and confirm your conversation survives a restart.
Choose software Run its commands Check the result
01
BEFORE YOU BEGIN
The software
Install the software for your operating system, then follow the steps to start your local server.
CURRENT SETUP
View software source Open WebUI + local Ollama · official
Follow the software’s documented system requirements.
02
THE WORKFLOW
Set up Open WebUI + local Ollama on your machine
Choose where you want to run it. The instructions below update to match your selection.
CHOOSE YOUR SETUPWindows
Local browser server with accounts and persistent storage. Keep both Ollama and Open WebUI running during use.
Install uv · PowerShell
Install uv from its official instructions, then open a new PowerShell window. The commands below create an isolated Python 3.11 environment; there is no need to change your system Python.
uv installationpowershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"Start the private web server
Run this in PowerShell and leave it open. It installs the selected Open WebUI release and Python 3.11. The explicit DATA_DIR keeps history outside the temporary package cache. Open http://127.0.0.1:8080 after the server reports it is ready.
Official Open WebUI quick start$env:DATA_DIR="$env:USERPROFILE\yourunai-webui-data"
$env:OLLAMA_BASE_URL='http://127.0.0.1:11434'
uvx --python 3.11 open-webui@0.11.4 serve --host 127.0.0.1 --port 8080Create the local administrator account
Open the local address, create your first account, and sign in. This first account manages this local Open WebUI installation; it is separate from your YouRunAI login. Keep authentication enabled. The guide binds to 127.0.0.1, so it is accessible only on this computer.
Connect to Ollama and select the model
Open Admin Panel → Settings → Connections. Confirm the Ollama connection is enabled and its URL is http://127.0.0.1:11434. Save and refresh the model list. Start a new chat and select qwen3.5:4b. Cloud/API providers are not needed for this setup.
Ollama connection setupSet a practical context and chat
In the chat controls, expand advanced parameters and set the Ollama context size (num_ctx) to 4096. Send the prompt “Remember my project is Cedar. Give me three names for a private notes app.” Then ask “What is my project called?” Check that the answer remembers Cedar. Leave the conversation open.
Ollama context and memorySave a conversation and restart
Give the chat a recognizable title such as Cedar test. Press Ctrl+C in the Open WebUI server window to stop it. Run the same launch command again, reopen http://127.0.0.1:8080, sign in, and confirm Cedar test is still in the sidebar with both messages. DATA_DIR must remain the same.
$env:DATA_DIR="$env:USERPROFILE\yourunai-webui-data"
$env:OLLAMA_BASE_URL='http://127.0.0.1:11434'
uvx --python 3.11 open-webui@0.11.4 serve --host 127.0.0.1 --port 8080Back up before an update
Stop Open WebUI, then copy the entire yourunai-webui-data folder to your backup location. Keep the original folder for normal restarts. Update deliberately using the official instructions and keep one server process during database migration. This example does not expose the server to the internet or configure a shared LAN service.
Updating Open WebUI03
SUCCESS CHECK · Open WebUI + local Ollama
Your private chat survives a restart
You can sign in locally, chat with qwen3.5:4b, and reopen the same saved conversation after stopping and restarting the web server.
http://127.0.0.1:8080Opens on your machineWhen it doesn’t go to plan
No model appears
Keep Ollama running, check ollama list, and confirm its connection URL in Open WebUI is http://127.0.0.1:11434. Pull qwen3.5:4b if it is missing, then refresh the model list.
Port 8080 is already in use
Stop the other local process or change --port 8080 to --port 8081 in the launch command and open the new port. Do not start two Open WebUI processes against the same data directory.
Chats disappear after restarting
Stop the server and relaunch with the same DATA_DIR value from this guide. A different directory creates a different local installation; your original folder still contains the previous history.
Cannot install Open WebUI
Use uvx --python 3.11 as shown. Do not install this release with Python 3.13; its supported Python range is 3.11–3.12.
REFERENCE LIBRARY
6 SOURCESSources and files
Official software, installation instructions, and support resources behind this setup.
Keep exploring
Text
SETUP GUIDEQwen3.5 · 9B private chat setup
Source-backed package and platform-route plan for private local chat with Qwen3.5 · 9B. Preserves the official Ollama 9B package, a smaller 4B family fallback, and a clearly labeled community uncensored GGUF option.
Freshness 41 out of 100. Release age: 230 days old, Feb 27, 2026. Measures age, not quality.
Popularity 93 out of 100. 8,591,266 downloads in 30 days; 2,171 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.
SOFTWAREOllama
Coding
SETUP GUIDEBuild a private coding assistant with Qwen3.8 27B
Run Qwen3.8 27B with Ollama, connect it to Cline in VS Code, and make one reviewable code change in a tiny local project.
Freshness 76 out of 100. Repository age: 71 days old, Aug 5, 2026. Measures age, not quality.
Popularity 95 out of 100. 7,363,968 downloads in 30 days; 17,749 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.
Arena Open Source Web coding: rank 10 of 55. Ranked model: qwen3.8-27b. Original model ranking, not a benchmark of this package. Export dated Oct 13, 2026. Source checked Oct 15, 2026. Human preference ranking, not hardware suitability.+3
SOFTWAREOllama + Visual Studio Code + Cline + Python
Images
SETUP GUIDECreate images locally with FLUX.2 Klein
Generate images from text prompts in ComfyUI with the fast distilled or more adjustable base 4B model. Save your results and reuse the visual workflow.
Freshness 35 out of 100. Release age: 273 days old, Jan 15, 2026. Measures age, not quality.
Popularity 76 out of 100. 423,736 downloads in 30 days; 1,043 lifetime likes. Checked Oct 15, 2026. Measures adoption, not quality.
SOFTWAREComfyUI