How To Setup Ollama LLM AI Chatbot With Open WebUI, An Open-source AI User Interface On Ubuntu VPS
47In this video, you will learn how to create your own self-hosted Ollama Large Language Model (LLM) Artificial Intelligence (AI) Chatbot or AI assistant with an Open WebUI User Interface (UI) also known as a dashboard running on a DigitalOcean Ubuntu Linux droplet. Droplets are what DigitalOcean calls its cloud servers which also go by Virtual Private Servers (VPS). Both Ollama and Open WebUI are Open-source and both can be installed with one click using DigitalOceans marketplace apps.
🔵 Free $200 DigitalOcean cloud credits using my referral link: https://digitalocean.pxf.io/c/1245219/1373759/15890
How To Get $200 In Free DigitalOcean Cloud Credits • How To Get $200 In Free DigitalOcean ...
Open WebUI (Models, Prompts, Tools, Functions) https://openwebui.com/
Ollama Models https://ollama.com/search
Steps To Make An Ollama and Open WebUI Server On DigitalOcean:
1. Go to https://digitalocean.pxf.io/c/1245219/1373759/15890 and create a free DigitalOcean account. This is my referral link and it will grant you $200 in free cloud credit for 60 days as a new user
2. Once you have created your DigitalOcean account, you will be taken to your DigitalOcean project dashboard. Click Create.
3. Click Droplets
5. Choose your droplets region
6. Under choose an image, click Marketplace
7. Search for Ollama with Open WebUI and click on it to select this image
8 Configure your droplets remaining settings to your liking and click on Create droplet
9. Once your droplet is up and running, copy its IP address and open up another tab on your browser
10. Type http:// and then paste in your droplets IP address and press Enter on your Keyboard like so:
http://YourDropletsIpAddress
11. Click Sign up to sign up to your self-hosted Open WebUI instance and create an account
12. Once logged in, you can interact with the preinstalled model by clicking on Select a model
13. Choose tinyllama:latest and send a message in the textbox at the bottom to the model to elicit a response
14. To add users you can click on your Open WebUI profile and click Admin Panel
15. Click + to Add User. Choose a role for that user, configure the remaining options and click Submit. New users that want to use your Open WebUI self-hosted instance can sign up and you can change their status from pending to user. This will same you time as you won't need to configure all the options for them
16. To add a new model to your Open WebUI instance made by the OpenWebUI community , click Workspace
17. Click Discover a model. A new tab will open taking you to https://openwebui.com/#open-webui-community
18. Click Sign in, followed by Create one to Sign up to Open WebUI Community. Click Unlock
Note: The Open WebUI Community platform account you make, does not sync with your self-hosted Open WebUI instance, and vice versa. They are two separate accounts independent of each other.
19. Search for your desired model to add, and click Get. In this video we demonstrate adding a character AI model
20. Enter your Open WebUI self-hosted instances URL address which will be http://YourDropletsIpAddress
21. Click Import to WebUI
22. Select a Base Model eg. tinyllama:latest
23. Click Save & Create
24. To interact with with your newly added model click on New Chat at the top left and select your new model from the drop down list, followed by your message
25. To add a new base model, click on your profile picture
26. Click Admin Panel
27. Click Settings
28. Click Models
29. Click on the hyperlinked text under the textbox which says enter model tag or navigate to https://ollama.com/library
30. Select a lightweight parameter count for your new base model, in this video demonstration qwen2:0.5b was chosen and the following model tag: ollama run qwen2:0.5b was copied and pasted into the relevant Models textbox on the Open WebUI instance
31. Click the download button to begin pulling the base model to your instance
32. Once the model was successfully added to your self-hosted Open WebUI instance, you can interact with it by clicking on New Chat, selecting the new model and then send your desired message.
33. You can delete a chat with a model by clicking on the three horizontal dots (More) next to the models chat you want to delete, click Delete followed by Confirm.
Congratulations, you have successfully created your own self-hosted Ollama AI chatbot with an Open WebUI dashboard using DigitalOceans 1-Click droplets.
Timestamps:
0:00 - Intro & Context
1:06 - Create Ollama With Open WebUI Droplet
3:47 - Communicate With An Ollama Model On Open WebUI
5:43 - Add Users To Open WebUI
6:44 - Get & Use A Model From Open WebUI Community Platform
10:16 - Download And Use A Base Model From Ollama
12:35 - Delete A Conversation With A Model
12:50 - Closing & Outro
More DigitalOcean Guides (Playlist) • DigitalOcean Cloud Computing Guides
AI, ML, & LLMs (Playlist) • Artificial Intelligence (AI), Machine...
#Ollama #OpenWebUI #DigitalOcean