Using 123API in ChatBox

Configure ChatBox to use 123API as a custom OpenAI-compatible endpoint.

Prerequisites

Install ChatBox
Generate an API key in the 123API Dashboard
Decide which model you want to use for daily conversations

Step 1: Open ChatBox Settings

Launch ChatBox
Open the setup wizard or click the settings icon
Choose to use your own API provider

Step 2: Configure Provider

In most cases, choose the OpenAI-compatible or OpenAI API option and fill in:

Field	Value
Provider	`OpenAI API`
API Host / Base URL	`https://123api.co/v1`
API Key	Your 123API key
Model	Example: `gpt-4o`

Some ChatBox versions ask for API Host instead of Base URL. If that field expects only the host, try https://123api.co first. If requests fail, switch to https://123api.co/v1.

Step 3: Choose a Model

Recommended starting set:

gpt-4o-mini for everyday use
gpt-4o for higher-quality output
gpt-5 for difficult reasoning prompts
claude-sonnet if your ChatBox build supports that route cleanly
gemini-2.0-flash for fast multimodal experimentation

Step 4: Start Using

Return to the main chat screen
Select the configured model
Send a small test prompt first
Confirm the response is normal before saving the configuration as default

Usage Tips

Keep one fast model for daily use and one stronger model for difficult prompts
If you switch model IDs manually, verify the exact name in GET /v1/models
Save long conversations on your own side if they are important business records

FAQ

ChatBox cannot connect+

Check the host value, confirm the API key is valid and test whether your local environment can access https://123api.co.

The model does not appear in the dropdown+

Type the model ID manually if ChatBox does not auto-refresh the list, and make sure the ID exists in GET /v1/models.

Responses fail after changing provider+

Re-open settings and verify that the provider type, host format and model family still match each other.

Open API Manual

Review the chat completion request format used by ChatBox-compatible integrations.