Skip to content
Updates & support on Telegram
UncensoredAI

~/.config/opencode/opencode.json

Native [OI] support

OpenCode with a custom [OI]-compatible provider

OpenCode supports custom providers through @ai-sdk/openai-compatible. Add one provider block, list the catalog ids you want, and switch models per session.

Base URL

https://uncensoredai.dev/v1

  1. 01

    Create the config file

    OpenCode reads opencode.json or opencode.jsonc from the project root, or from the global config directory. The provider block below registers UncensoredAI.

    opencode.json
    {
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "uncensoredai": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "UncensoredAI",
      "options": {
        "baseURL": "https://uncensoredai.dev/v1",
        "apiKey": "{env:UNCENSOREDAI_API_KEY}"
      },
      "models": {
        "deepseek-v4.1-flash": { "name": "DeepSeek V4.1 Flash" },
        "gpt-6-sol": { "name": "GPT-6 Sol" },
        "glm-5.3-flash-uncensored": { "name": "GLM-5.3 Flash Uncensored" }
      }
    }
  },
  "model": "uncensoredai/deepseek-v4.1-flash"
}
  2. 02

    Set the key

    Keep the key out of the config file by using the env placeholder, then export it.

    macOS / Linux
    export UNCENSOREDAI_API_KEY="sk-ua-XXXX-XXXX-XXXX-XXXX"
    Windows (PowerShell)
    $env:UNCENSOREDAI_API_KEY = "sk-ua-XXXX-XXXX-XXXX-XXXX"
  3. 03

    Pick the provider in the model list

    Restart OpenCode and select a model under UncensoredAI. Model ids must match the catalog exactly; OpenCode does not fetch /v1/models automatically.

FAQ

Does OpenCode support arbitrary [OI]-compatible endpoints?+

Yes. The @ai-sdk/openai-compatible provider takes a base URL and API key, so any chat-completions-compatible gateway works.

Why is my model missing from OpenCode?+

The models map must list each id you want; OpenCode does not discover them. Copy exact ids from the model catalog.

More integrations

Generate access keyAll integrations