Connecting LangBot

Difficulty 🧩 Medium　|　API 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

LangBot is an open-source platform for building instant-messaging bots. It supports many platforms such as Discord, Telegram, Slack, Feishu, DingTalk, WeCom and QQ, and comes with built-in knowledge base, Agent and MCP capabilities. Point it at byesu — an AI API gateway with OpenAI-compatible and Anthropic-native endpoints — and one key powers your chatbot with Claude, GPT-5.6, Grok, Gemini and more, billed pay-as-you-go with no subscription. LangBot uses the OpenAI-compatible endpoint, so all it needs is a custom base URL and a key.

Before you start

First go to Console → Tokens and create an API token (it looks like sk-xxxx).

When you create it, be sure to pick the right group — the group decides which models you can call and which route they go through. Not sure how to choose? See Choosing a group.

Step 1: Install and start LangBot

LangBot offers several deployment methods; the recommended one is a one-command Docker start.

Docker (recommended)

git clone https://github.com/langbot-app/LangBot
cd LangBot
docker compose up -d

Manual deployment

# Download the installer for your platform from Releases
# https://github.com/langbot-app/LangBot/releases
# Unzip it and start it as described in the docs

After it starts, open http://localhost:5300 in your browser, follow the prompts to finish setup (set the admin username and password), and enter the admin dashboard.

Official documentation: https://docs.langbot.app

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

In the LangBot admin dashboard, open the Models management page on the left, click Add model, and fill in the fields as shown below:

Field	What to enter
Model provider / Request URL	https://byesu.com/v1
API Key	Your token sk-xxxx
Model name	gpt-5.5 (or claude-opus-4-8, gemini-3.1-pro-high)

The base_url must include /v1

LangBot uses the OpenAI-compatible interface, so the request URL must be https://byesu.com/v1, ending with /v1. If you leave it off, you'll get a 404 error.

If the provider list has a NewAPI option, just pick it and change the request URL to https://byesu.com/v1. If it doesn't, pick the OpenAI-compatible provider instead — the result is the same.

The model name must match one that's actually available in your group. Common choices:

• Chat scenarios: gpt-5.5, claude-opus-4-8
• Long text / reasoning: gemini-3.1-pro-high

Step 3: Enable it in the pipeline and verify

1. Go to the Pipeline configuration, and in the model selector of the conversation node, choose the model you added in Step 2.
2. After saving, send a message in the built-in conversation debug window to test it.
3. If you get a normal reply, the connection is working. Then bind your Discord / Telegram or other platform accounts to go live.

Knowledge base (optional)

LangBot supports building a vector knowledge base with an Embedding model. Add another model, set the request URL to https://byesu.com/v1 as well, set the model name to an Embedding model available in your group, and then select it when you create a new knowledge base.

Stuck?

• 401 / authentication failed: The API Key is wrong or has extra spaces. Go back to the Console and copy the complete sk-xxxx again. See Common errors #auth for details.
• 404 / endpoint not found: The request URL is missing /v1; it should be https://byesu.com/v1.
• No available channel / model does not exist: Your token's group doesn't include this model. Try a different model name, or recreate the token with the right group. See Choosing a group and Common errors #no-channel.
• Insufficient balance: Your account credit is used up. Top up and try again. See Common errors #balance.

For more troubleshooting, see Common errors.