Connect an app
Open WebUI
Add Kyara Intelligence to Open WebUI as an OpenAI API connection, and its models appear in the model picker.
Open WebUI connects to Kyara through its built-in OpenAI API connection. Add one connection with the Kyara base URL and your API key, and the models from the catalog show up in the model picker. No plugins needed.
Have your API key from the dashboard ready before you start.
| Setting | Value |
|---|---|
| URL | https://api.kyara-intelligence.com/v1 |
| Key | Your key from the dashboard |
Add the connection
Open the admin settings
Click your profile icon, open the Admin Panel, then go to Settings and Connections.
Add an OpenAI API connection
Under OpenAI API, click + to add a connection.
Enter the base URL and key
Set the URL to:
https://api.kyara-intelligence.com/v1Then paste your Kyara API key into the Key field.
Save
Click Save. Open WebUI loads the model list from
GET /v1/models, so the catalog is available right away.Start chatting
Open a new chat, pick a model such as
deepseek/deepseek-v4-flashfrom the selector at the top, and send a message.
Troubleshooting
No models show up
Reopen the connection and check that the URL ends in /v1 with nothing after it, and that the key was
saved. Then reload the page so the model list loads again.
Requests fail with a 401
The key is missing, mistyped, or was regenerated. Copy a fresh key from the dashboard, paste it into the connection's Key field, and save again.
A feature isn't working
Kyara is text-only, so image input and native tool calling are not available. See Unsupported features for the full list.
Check your key first
To confirm your credentials before touching Open WebUI, export your key with export KYARA_INTELLIGENCE_API_KEY="your-api-key", then send this request from the same terminal. A
response with a choices array means the key works; an error with invalid_api_key means it is missing or wrong.
curl https://api.kyara-intelligence.com/v1/chat/completions \
-H "Authorization: Bearer $KYARA_INTELLIGENCE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v4-flash",
"messages": [
{ "role": "user", "content": "Hello!" }
]
}'