Horosa.app中文

AI Setup Guide

Configure the AI Key from this site in the Horosa client, then use the AI analysis module with mainstream models for chart notes, references, and structured interpretation.

Before you configure

Use the Key only inside your own Horosa client. This site does not store your Key. Keep it private and do not share screenshots or messages containing it, or the points may be used by others.

Base URL
https://oa.ai01.org/v1
1

Open the AI analysis module

Click the current module name in the top bar, open “Select function module”, then choose “AI Analysis” under the tools section.

Open the AI analysis module
2

Open settings and add an endpoint

Inside AI Analysis, click “Settings” on the right, then click “Add interface configuration” at the top. On first setup, the list may be empty.

Open settings and add an endpoint
3

Fill in the Key and Base URL

Use the custom OpenAI-compatible endpoint preset, paste your Key into API Key, enter Base URL: https://oa.ai01.org/v1, and keep this configuration enabled.

Fill in the Key and Base URL
4

Pull the model list

After saving the endpoint, return to the endpoint card and click “Pull models”. The left screenshot shows the unchecked state; the right screenshot shows models after a successful pull.

Pull the model list 1Pull the model list 2
5

Choose models and start analysis

Return to the Analysis tab, select the chat model, embedding model, chart, and analysis type. Once the connection test succeeds, enter your question and send it.

Choose models and start analysis

Recommended settings

  • For chat models, start with high-capability models such as GPT-5.5, Claude Opus 4.8, or gemini-3.1-pro.
  • Embedding models are used for reference retrieval and context matching. Choose an available model returned by the client.
  • If connection fails, check that the Key is complete and the Base URL is exactly https://oa.ai01.org/v1, then test again.