Cherry Studio with gpt88.cc

This guide walks the full setup path: opening provider settings, selecting the right route, verifying connectivity, and debugging failures without guessing.

What this guide covers

• How to connect Cherry Studio to the gpt88.cc OpenAI-compatible API.
• How to start with one stable model and expand later.
• What to check first when the connection fails.

Preparation

1. Set the provider to OpenAI Compatible
2. Make sure Base URL is exactly https://api.gpt88.cc
3. Use a real API key from the console
4. Start with a known stable chat model
5. Send one minimal message to verify
6. If it fails, validate with curl first

Quick setup

1. Open Cherry Studio
2. Go to model provider / API settings
3. Choose OpenAI Compatible
4. Set Base URL to https://api.gpt88.cc
5. Paste the API key from the gpt88.cc console
6. Start with one stable chat model
7. Send one minimal test message

Use one minimal request to confirm the path first, then import additional models or prompt templates.

curl https://api.gpt88.cc/v1/chat/completions \
  -H "Authorization: Bearer $GPT88_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "messages": [{"role":"user","content":"Introduce gpt88.cc"}]
  }'

Verify connectivity

1. Save the Cherry Studio provider settings.
2. Create a new chat and send one short prompt.
3. If that works, then gradually increase prompt or context complexity.
4. If it fails, verify the API key and Base URL with curl first.

Usage tips

• Pin one default model to reduce unnecessary switching.
• Use stronger models for long-context tasks and cheaper models for fast Q&A.
• Create separate API keys per project for cost tracking and easier revocation.

Common questions

• If the model list does not load, enter the model ID manually.
• If you see 404, check whether /v1 is present in the Base URL.
• If you see 401, verify that the API key was copied completely.

Troubleshooting

1. Model list is blank
   - Enter the model ID manually
   - Copy the exact name from the model catalog

2. 401 response
   - Key is invalid or expired

3. 404 response
   - Base URL is wrong or /v1 is missing

4. Replies are slow
   - Switch to a lighter model first
   - Disable knowledge base and long multi-turn context for the first test

Next steps

• Return to the integrations hub.
• Read the complete integration guide.