Connect OpenClaw to a paid LLM API.

Add Yolo-Auto as an OpenAI-compatible OpenClaw provider and use Qwen3.8-27B with a flat-rate paid plan.

Create account → Open full OpenClaw docs

Agent-ready config

Copy the provider block into the selected OpenClaw agent's models.json.

Paid inference

Free-tier requests are temporarily disabled. Paid plans are available now to test model inference.

Flat-rate upgrade

Keep the same provider configuration when usage grows.

Quick setup

Create an account, choose a paid plan, copy your yolo_... API key, set your base URL to https://yolo-auto.com/v1, and select qwen3.8-27b for paid inference. See the models page for details.

Best next pages

Docs · Pricing · Qwen3.8-27B API · LLM API access · Flat-rate LLM API

Configure the selected OpenClaw agent

OpenClaw keeps model configuration under ~/.openclaw/agents/. Update the models.json file for the agent you run, then select the Yolo-Auto Qwen3.8-27B model.

Use a custom OpenAI-compatible provider

The provider block supplies the Yolo-Auto base URL, API key, API mode, model ID, context window, and reasoning support expected by the agent.

Keep your configuration across paid plans

Paid plans use the same endpoint and key shape, so changing your paid plan does not require a new integration.

Who OpenClaw LLM API Setup is actually for

OpenClaw LLM API Setup is best for developers and power users who want model access inside tools, agents, scripts, and apps, not just a closed consumer chatbot tab.

Keep the coding agent, change the model backend

The integration uses each tool's existing custom-provider support. Yolo-Auto supplies an OpenAI-compatible endpoint, API key, and public Qwen3.8-27B model without replacing the coding-agent interface.

Choose a paid plan to verify the configuration with inference requests. The same endpoint and model shape continue working when a flat-rate paid plan becomes a better fit for longer sessions.

Add Yolo-Auto to the agent's models.json

Copy the provider block, replace the API-key placeholder, and save it in the configuration for the OpenClaw agent you run.

These examples use Qwen3.8-27B with model ID qwen3.8-27b for text, images, and tools on paid plans. Paid users can optionally choose yolo, a configurable route currently backed by Qwen3.8-27B. Its server-side target can change while your client model ID stays the same.

{
  "providers": {
    "yolo-auto": {
      "baseUrl": "https://yolo-auto.com/v1",
      "apiKey": "ENTER_YOUR_yolo_KEY_HERE",
      "api": "openai-completions",
      "models": [
        {
          "id": "qwen3.8-27b",
          "name": "Qwen3.8-27B",
          "contextWindow": 128000,
          "reasoning": true,
          "input": [
            "text",
            "image"
          ]
        }
      ]
    }
  }
}

Keep the configuration private because it contains the API key.

When to choose Yolo-Auto

Choose it when

You need OpenAI-compatible LLM access, predictable cost, no prompt retention. Your prompts and responses are not used for model training.

Skip it when

You need image generation, every model under the sun, a managed IDE, or a consumer-only chatbot with no API workflow.

Next step

Read the docs, check models, compare pricing, or review the privacy policy.

OpenClaw LLM API Setup FAQ

Can OpenClaw use Yolo-Auto?

Yes. Add Yolo-Auto as an OpenAI-compatible provider in the selected agent's models.json file.

Where does the OpenClaw API key go?

Place the Yolo-Auto key in the provider configuration shown below and keep that file out of public repositories.

Can I test OpenClaw for free?

Free-tier requests are temporarily disabled. Paid plans offer OpenClaw inference through Yolo-Auto now.

Related LLM API guides