📖 查看中文版 →
Prerequisites #
- Obsidian installed — obsidian.md
- Obsidian Copilot plugin installed and enabled
- A WoAiToken account with credits — to.woaitoken.com
- Your sub2api API key (get it in Step 1)
> This guide uses the Obsidian Copilot plugin. If you use a different plugin (Smart Connections, Text Generator, etc.), the settings location will differ but the API Base URL and Key are the same.
Step 1 — Get Your API Key #
1. Open the console
Visit to.woaitoken.com/console and sign in.
2. Create a token
Go to 令牌 (Tokens) → Add Token. Give it a name (e.g. "Obsidian"), set no expiration, then click Submit.
3. Copy the key
Copy the generated key — it starts with sk-. Save it immediately; it is only shown once.
Step 2 — Install Obsidian Copilot #
1. Open Obsidian → Settings → Community plugins 2. Click Browse, search for Copilot 3. Install and enable the plugin
Step 3 — Configure Obsidian Copilot #
1. Go to Settings → Copilot 2. Under Chat Model, set Model Provider to openai 3. Enter your OpenAI API Key: sk-your-key-here 4. Scroll down to Advanced Settings 5. Enable Use a proxy/custom endpoint 6. Set OpenAI API Base to: https://to.woaitoken.com/v1 7. Choose a Model (e.g. claude-sonnet-4-5 or gpt-4o) 8. Click Save
> 📸 Add screenshot of Copilot settings panel here.
Step 4 — Verify #
1. Open the Copilot panel (Ctrl/Cmd + Shift + P → search "Copilot") 2. Type a test message (e.g. "Hello") 3. A response confirms the connection is working
Troubleshooting #
"Invalid API key" error Make sure the key starts with sk- and has no extra spaces.
No response / timeout Confirm the API Base URL is https://to.woaitoken.com/v1 (with /v1).
Model not available Check available models at to.woaitoken.com/console. Use only models listed there.
429 Too Many Requests Your credits are used up. Top up at woaitoken.com.
