Use ApexApi in OpenCode

OpenCode is an open-source coding agent for the terminal and desktop. Add ApexApi as a custom provider once and every model you list from our catalog shows up in its model picker.

1. Get your API key

Create a key in your dashboard. It starts with ak-.

2. Export your key

export APEXAPI_API_KEY=ak-your-key-here

3. Add ApexApi to opencode.json

Edit ~/.config/opencode/opencode.json (or a project-level opencode.json) and add:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "apexapi": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "ApexApi",
      "options": {
        "baseURL": "https://api.apexapi.dev/v1",
        "apiKey": "{env:APEXAPI_API_KEY}"
      },
      "models": {
        "anthropic/claude-sonnet-4.6": { "name": "Claude Sonnet 4.6" },
        "openai/gpt-5.6-sol": { "name": "GPT 5.6 Sol" },
        "deepseek/deepseek-v4-pro": { "name": "DeepSeek V4 Pro" }
      }
    }
  }
}

The ID must match the catalog exactly, in provider/model format. Browse all IDs on the models page.

4. Select a model

Start OpenCode and run /models; the ApexApi entries are in the list. The same config powers OpenCode Desktop, which reads opencode.json too; on desktop you pick the model from the dropdown instead of the command.

Troubleshooting

401 Unauthorized: the key wasn't pasted correctly or has been revoked.
Model not found: the model ID doesn't match a catalog ID. Copy it verbatim from the models page.
Connection fails: make sure the base URL is exactly https://api.apexapi.dev/v1 (no trailing slash after v1) and your balance is positive.
Provider missing from /models: the JSON didn't parse or the env variable isn't set in the shell that launched OpenCode. Validate the file and run echo $APEXAPI_API_KEY.
Anything else: check the error codes reference; streaming and tool calling are fully supported on /v1/chat/completions.