Tutorial Collection — 300+ AI Agents & Coze (English edition)

Step 1. Go to the official site and register an account (Required reading)

Step 1. After registering, use a redemption code to redeem credits (you can also top up online here for renewal)

Step 2. Then create a Key — just fill in a name under API Key; the group field is optional

Step 3. Copy the created token key

Step 4. You can now configure it wherever you want — e.g. Claude Code, CC Switch, Codex, etc. (When configuring, choose the OPENAI protocol — custom model configuration)

OpenAI protocol endpoint: https://www.yundare.com/v1

Anthropic protocol endpoint: https://www.yundare.com

You only need three things: the endpoint + the token + the model name you want to use (check the "Model Square" at https://www.yundare.com/pricing to see which models are available, then copy its name)

Can't connect? How to quickly verify?

Go to the Playground at https://www.yundare.com/playground and select a model. If the model responds normally, the issue is in your client configuration. Otherwise it's a model-routing problem — try another model or contact us for feedback.

2. Claude Code Configuration

Step 1. First download the latest version of CC Switch (not required, but makes life easier) — CC Switch download URL

[CC-Switch-v3.17.0-Windows.msi]

Step 2. Go to https://www.yundare.com/keys — API Key Management

Select / fill in the relevant connection info

Import the configuration into CC Switch

Step 3. After import, click Test and check the status

Step 4. Open the terminal

Step 5. The Claude Code graphical plugin for VS Code also works (remember to restart VS Code)

Step 6. Configuration issues

Issue 1

(If you encounter this issue, it's because the Claude plugin doesn't support the OpenAI protocol — see image 2 to set the OpenAI protocol and enable routing)

Change the API format to OpenAI protocol

Enable routing settings and turn on Claude routing

Issue 2

Step 1. Update to the latest version.

Step 2. Switch to Codex's native protocol (no routing needed)

Step 3. Quit Codex completely — end the process, then reopen it

3. Codex Configuration

Step 1. First download CC Switch (not required, but makes life easier) — CC Switch download URL

Codex download URL

Step 2. Go to https://www.yundare.com/keys — API Key Management

Select / fill in the relevant connection info

Import the configuration into CC Switch

Step 4. After import, click Test and check the status

Step 5. Restart Codex to start chatting. [If you're not configuring a GPT-series model, the model field will be blank, but it still works.] [The blank one calls the first model you selected.]

[You need to restart Codex every time you switch models in CC]

3. Use in Roo Code / Kilo Code

API provider: select Anthropic

Fill in the API Key

Use a custom Base URL and enter https://www.yundare.com/

4. Use in Cherry Studio

[Cherry Studio.exe]

Add a new provider

Type: select Anthropic

Fill in API Key and API URL https://www.yundare.com/

Click Manage to fetch models and add them

5. OpenClaw Model Configuration

You need three things — Base URL: https://www.yundare.com/v1

Key: (the key obtained in the Key creation step). Model ID: (the model name under the selected group)

# Edit the config file ~/.openclaw/openclaw.json
# Find baseUrl, apiKey, and Model ID, replace them, then save
{
  "models": {
    "providers": {
      "aiii": {
        "baseUrl": "https://www.yundare.com/v1",
        "apiKey": "sk-your-api-key",
        "api": "openai-completions",
        "models": [
          {
            "id": "gpt-5.6-luna",
            "name": "gpt-5.6-luna",
            "contextWindow": 1048576
          }
        ]
      }
    }
  }
}

# Restart the gateway to apply the config
openclaw gateway restart

# Check whether the model was added successfully
openclaw model list

Codex routing beginner tutorial!: Low-priced tokens at https://www.yundare.com/

Coze workflow… files shared via cloud drive:

Link: https://pan.baidu.com/s/1QbZSmzhOrXp-TKa2-DPoPw?pwd=7dgj