OpenCode

Use TensorX models with OpenCode, the open-source terminal-based AI coding agent.

---

Prerequisites

• OpenCode installed

• A TensorX API key from app.tensorx.ai

---

Configuration

Step 1: Connect the Provider

Run /connect in OpenCode, select Other, and enter:

Field	Value
Provider ID	tensorx
API Key	Your TensorX API key

Step 2: Configure opencode.json

Add the following to your project's opencode.json (or ~/.opencode/opencode.json for global config):

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "tensorx": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "TensorX",
      "options": {
        "baseURL": "https://api.tensorx.ai/v1",
        "apiKey": "{env:TENSORX_API_KEY}"
      },
      "models": {
        "<MODEL_ID>": {
          "name": "TensorX model"
        }
      }
    }
  }
}

Replace <MODEL_ID> with an ID returned by GET /v1/models. Add optional limits only after reading the selected model's metadata from GET /v1/model/info.

Step 3: Set API Key

Set your API key as an environment variable:

export TENSORX_API_KEY=your-tensorx-api-key-here

Or add to ~/.opencode/.env:

TENSORX_API_KEY=your-tensorx-api-key-here

---

Vision / Image Support

To use a model that accepts image input, add modalities to its model configuration. Without this setting, OpenCode strips images before sending them.

{
  "<VISION_MODEL_ID>": {
    "name": "TensorX vision model",
    "modalities": {
      "input": ["text", "image"],
      "output": ["text"]
    }
  }
}

---

Model selection

Call GET /v1/models with your TensorX API key and use one of the returned IDs. If the integration depends on a capability or token limit, query GET /v1/model/info before selecting the model. See Models for examples.

---

Tips

• For vision tasks, remember to add the modalities config shown above

• Use the model ID as the configuration key

Resources

• OpenCode Documentation

• TensorX API Reference

• TensorX Model Catalog