Horosa AI Key Setup Guide for Xingque AI
A practical setup guide for adding your AI Key, Base URL, model list, and AI analysis configuration inside the Horosa-based Xingque AI client.
Claim and store your AI Key first
The Xingque AI Key is anonymous and does not require an account. After claiming it, store it in your own password manager, private note, or local secure location.
Horosa.app does not store your Key. If the Key is leaked, other people may use the points inside it, so avoid posting the full Key in chats, screenshots, or public pages.
Add a compatible custom endpoint
Open the Xingque AI client, enter the AI analysis module, and add a new interface configuration in settings. Choose a custom compatible provider and paste your Key into the API Key field.
Use https://oa.ai01.org/v1 as the Base URL. Before saving, confirm that the configuration is enabled so the client can use it.
Fetch models and run a test
After saving the interface, fetch the model list. When the list is returned, you can prioritize capable models such as GPT-5.5, Claude Opus 4.8, and gemini-3.1-pro, or choose from the actual available list.
Return to the analysis page, select chat model, embedding model, chart, and analysis type. Once the test succeeds, you can use AI to help organize chart data, references, and reasoning notes.
Frequently Asked Questions
What should I use as the Base URL?
Use https://oa.ai01.org/v1 and make sure there are no extra spaces.
What if fetching models fails?
Check whether the Key is complete, the Base URL is correct, and the network is available. Then save the interface again and fetch models.