Integrations

opencode

Use TokenRouter as an OpenAI-compatible provider in opencode.

opencode supports any OpenAI-compatible provider. Add TokenRouter to your opencode.json and every model behind the gateway becomes available in the model picker.

Configure

json

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "tokenrouter": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "TokenRouter",
      "options": {
        "baseURL": "https://api.tokenrouter.io/v1",
        "apiKey": "{env:TOKENROUTER_API_KEY}"
      },
      "models": {
        "openai/gpt-5-mini": { "name": "GPT-5 mini" },
        "anthropic/claude-sonnet-4-5": { "name": "Claude Sonnet 4.5" },
        "deepseek/deepseek-reasoner": { "name": "DeepSeek Reasoner" },
        "auto": { "name": "TokenRouter Auto" }
      }
    }
  }
}

Run

bash

export TOKENROUTER_API_KEY=tr_your_key_here
opencode

Then pick a tokenrouter/... model from the model list.

Reasoning models

TokenRouter returns reasoning_content for DeepSeek-family reasoning models in both streaming and non-streaming chat completions, so opencode renders thinking traces correctly for models like deepseek/deepseek-reasoner.