OpenCode

Prerequisites

OpenCode installed
A Chuizi.AI account with an API key (starts with ck-)

Option 1: Anthropic Protocol (Recommended for Claude)

Configure via environment variables for Claude model users:

terminal

bash

export ANTHROPIC_BASE_URL=https://api.chuizi.ai/anthropic
export ANTHROPIC_API_KEY=ck-your-key-here

To make this permanent, add to your shell configuration:

terminal

bash

echo 'export ANTHROPIC_BASE_URL=https://api.chuizi.ai/anthropic' >> ~/.zshrc
echo 'export ANTHROPIC_API_KEY=ck-your-key-here' >> ~/.zshrc
source ~/.zshrc

Option 2: OpenAI-Compatible Protocol

Use this when working with GPT, DeepSeek, or other models:

terminal

bash

export OPENAI_BASE_URL=https://api.chuizi.ai/v1
export OPENAI_API_KEY=ck-your-key-here

To make this permanent:

terminal

bash

echo 'export OPENAI_BASE_URL=https://api.chuizi.ai/v1' >> ~/.zshrc
echo 'export OPENAI_API_KEY=ck-your-key-here' >> ~/.zshrc
source ~/.zshrc

Usage

Once configured, launch OpenCode in your project directory:

terminal

bash

cd your-project
opencode

OpenCode will automatically detect the environment variables and connect to Chuizi.AI.

Verify the Configuration

Run opencode in a project directory
Send a test message, such as "describe this project"
If you receive a normal response, the configuration is working
Check the Chuizi.AI dashboard for request logs

Common Issues

Environment Variables Not Taking Effect

Make sure you have run source ~/.zshrc or opened a new terminal window. Run echo $ANTHROPIC_BASE_URL to verify the variable is set correctly.

Both Protocols Configured

If you have both Anthropic and OpenAI environment variables set, OpenCode will automatically use the appropriate protocol based on your selected model. You can configure both, but it is simpler to set up only the one you use most.

Rate Limiting

Connecting through Chuizi.AI reduces provider-specific interruptions through managed routing and capacity. If you encounter 429 errors, check your key's rate limit settings in the Chuizi.AI dashboard.

Next Steps

Choose a Model — find the best model for your coding tasks
Choosing a Protocol — OpenAI vs Anthropic protocol trade-offs
Cost Optimization — reduce costs with model tiering and caching