Set up OpenCode
This guide uses GPT-6.1 Sol (model ID gpt-6.1-sol). Copy your I Code Easy API key from API keys and choose the Standard endpoint for GPT models in Endpoint selection.
OpenCode
If you opened this guide from a model page, replace the example gpt-6.1-sol with the model ID on the card and use the protocol and endpoint indicated there.
Merge this configuration into your project’s opencode.json or global ~/.config/opencode/opencode.json. Set ICODEEASY_API_KEY in the local environment, then start OpenCode from that environment. Select GPT-6.1 Sol under I Code Easy in the model picker.
{
"$schema": "https://opencode.ai/config.json",
"model": "icodeeasy/gpt-6.1-sol",
"providers": {
"icodeeasy": {
"name": "I Code Easy",
"env": [
"ICODEEASY_API_KEY"
],
"package": "@opencode/ai/providers/openai-compatible",
"settings": {
"baseURL": "https://api.icodeeasy.cc/v1"
},
"models": {
"gpt-6.1-sol": {
"name": "GPT-6.1 Sol"
}
}
}
}
}
For another model, replace the ID after icodeeasy/ in model and the matching key in providers.icodeeasy.models. Keep the icodeeasy/ prefix. This example uses the OpenCode 2 config format.
Connect through Responses
Keep settings.baseURL and env from the OpenCode 2 example above and set package to @opencode/ai/providers/openai-compatible/responses. Use the same model ID in model and models.
Connect through Messages
Keep settings.baseURL and env from the OpenCode 2 example above and set package to @opencode/ai/providers/anthropic-compatible. Use the same model ID in model and models.
Verify your first request
Send a short text prompt in a new session and confirm the model and status in Usage logs. For a 401, check your key. For model or permission errors, check the model ID and the endpoint listed on the model page. If you have connection issues, try https://jp.icodeeasy.cc or https://sg.icodeeasy.cc with the same path.
Official client documentation
Related guides
Codex Setup
Install Codex CLI, store the API key in auth.json, configure the I Code Easy provider and Base URL, and start a first Codex session.
Chat Completions
Send a conversation using the OpenAI Chat format. Check that your model and selected endpoint support the Chat protocol before sending the request.
How do I troubleshoot slow connections or failed requests?
Match 401, 403, 429, timeouts, and Gemini path errors before changing hosts. Switch to the jp or sg backup domain only after the network is the problem.