AstrBot Integration

Difficulty ⌨️ Advanced　|　Interface format OpenAI compatible　|　base_url https://byesu.com/v1

Setup · fill these in

API addresshttps://byesu.com/v1

API keysk-xxxxxxxxCreate in console →

Modelgpt-5.5

API format OpenAI Compatible · model names follow the console

AstrBot is an open-source, all-in-one chatbot Agent platform that lets you bring AI into chat apps like QQ, Feishu, DingTalk, WeCom, and more. Point its "model provider" at byesu — an AI API gateway with OpenAI-compatible and Anthropic-native endpoints — and your bot can use Claude, GPT, Grok and Gemini models such as gpt-5.5 and claude-opus-4-8 with a single key, pay-as-you-go, no subscription.

Before you start

First go to Console → Tokens and create a token (something like sk-xxxx), then pick the right group for the token — the group determines which models you can call and how you're billed. If you're not sure how to choose, see Choosing a Group.

Step 1: Deploy and open the AstrBot admin panel

AstrBot supports several deployment methods, including Docker, BT Panel, and source code. For deployment details, see the AstrBot official documentation. One-line Docker startup example:

docker run -d \
  --name astrbot \
  -p 6185:6185 \
  -v $PWD/astrbot/data:/AstrBot/data \
  soulter/astrbot:latest

Once it's running, open the admin panel in your browser:

http://your-server-IP:6185

The first time you open it, you'll be guided through setting up an admin username and password. After logging in, you'll land in the console.

Step 2: Add a model provider (enter base_url + Key)

In the left sidebar of the admin panel, go to Model Provider (Provider), click Add / New Service Provider, and fill in the fields as follows:

Field	What to enter
Provider type	Choose OpenAI (OpenAI-compatible interface)
API Base URL	https://byesu.com/v1
API Key	Your token sk-xxxx
Model	A real byesu model name, such as gpt-5.5

The base_url must include /v1

AstrBot uses the OpenAI-compatible interface, so the API Base URL must be set to https://byesu.com/v1 (ending in /v1). Leaving off /v1 will cause requests to fail with 404 / unable to connect.

Common model names (fill in as needed; you can configure several in the panel):

gpt-5.5                # OpenAI, general-purpose chat
claude-opus-4-8        # Anthropic, strong reasoning / long-form
gemini-3.1-pro-high    # Google, multimodal

When you're done, click Save. Some versions provide a "Test / Fetch model list" button — you can click it to confirm the base_url and Key are connected.

Step 3: Set it as the default model and verify

1. Go to the Configuration → Model / Large Language Model area.
2. Set the Default Chat Provider / Model to the byesu provider you just created.
3. Click Save, and restart / reload the configuration if needed.

Then, on a chat platform you've connected (or via the panel's built-in chat test entry), send a message. If the bot replies normally, the integration is successful. If the reply looks wrong, go back to the provider configuration and double-check the base_url, Key, and model name.

Stuck?

• 401 / authentication failed: The Key is wrong or is missing the sk- prefix; copy the full token and re-enter it. See Common Errors · Authentication for details.
• 404 / can't connect: The API Base URL is missing /v1; it should be https://byesu.com/v1.
• "No available channel" / model does not exist: The token's group doesn't cover this model, or the model name is misspelled; switch to a real model name such as gpt-5.5 or claude-opus-4-8, and pick the right group. See Common Errors · No Available Channel.
• Insufficient quota: The account or token has run out of quota; top up (USDT, Alipay or WeChat Pay) or raise the token quota and try again. See Common Errors · Balance.

For more troubleshooting, see Common Errors.