Complete Guide: Set Up OpenClaw with NovAI in 5 Minutes

Create Your NovAI Account

Go to aiapi-pro.com.

1. Click "Get Started" (or "Register" in the navigation bar)
2. Enter your email address
3. Choose a password
4. Click "Register"

After registration, you'll be automatically logged in and taken to your Dashboard. This is your home base — you'll see your balance, API key, and usage stats here.

Add Balance (or Start Free)

Option A: Start 100% Free

If you just want to test things out, skip this step entirely. NovAI includes a completely free model — GLM-4.6V-Flash — with no usage limits and no balance required. Jump straight to Step 3.

Option B: Add Balance with USDT

To use DeepSeek, Qwen, MiniMax, and other paid models, you'll need to add balance. Here's how:

1. In your Dashboard, scroll down to the "Top Up" section
2. Choose a top-up amount:
   • $5 — Good for testing and light use (lasts 1-2 weeks)
   • $20 — Recommended for regular OpenClaw users (lasts 1-2 months)
   • $50 — Best value for heavy daily usage
3. Click the amount card → a payment method selection appears
4. Select "USDT (TRC20)"
5. A wallet address and QR code will be displayed
6. Send the exact USDT amount to this address on the TRC20 network
7. Wait for confirmation — your balance will update automatically (usually 1-3 minutes)

How Long Does $5 Last?

Get Your API Key

Your API key was generated automatically when you registered. Here's how to find it:

1. Go to your Dashboard (click the NovAI logo or "Dashboard" link)
2. Look for the "API Key" section near the top
3. Your key is hidden by default — click "Show" to reveal it
4. Click "Copy" to copy it to your clipboard

Your API key looks like this: nvai-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Store It Safely as an Environment Variable

The best practice is to store your key as an environment variable so it never appears in config files:

# Linux / macOS: add to ~/.bashrc or ~/.zshrc
export NOVAI_API_KEY="nvai-your-actual-key-here"

# Then reload your shell
source ~/.bashrc   # or: source ~/.zshrc

# Windows PowerShell: add to $PROFILE
$env:NOVAI_API_KEY = "nvai-your-actual-key-here"

# Or set permanently:
[System.Environment]::SetEnvironmentVariable("NOVAI_API_KEY", "nvai-your-actual-key-here", "User")

Quick Test: Verify Your Key Works

Before configuring OpenClaw, let's make sure the key works:

# Test with curl (using the free model — no balance needed)
curl https://aiapi-pro.com/v1/chat/completions \
  -H "Authorization: Bearer $NOVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-4.6v-flash",
    "messages": [{"role": "user", "content": "Say hello"}],
    "max_tokens": 50
  }'

If you get a JSON response with the model's reply, your key is working. If you get an authentication error, double-check that you copied the full key.

Configure OpenClaw

Now the main event. We'll add NovAI as a custom provider in OpenClaw.

Find Your Config File

OpenClaw's configuration file is located at:

# Linux / macOS
~/.openclaw/openclaw.json

# Windows
%USERPROFILE%\.openclaw\openclaw.json

If the file doesn't exist yet, create it. If it already has content, you'll merge the new config in.

The Complete Configuration

Add this to your openclaw.json:

{
  "models": {
    "mode": "merge",
    "providers": {
      "novai": {
        "baseUrl": "https://aiapi-pro.com/v1",
        "apiKey": "${NOVAI_API_KEY:-}",
        "api": "openai-completions",
        "models": [
          {
            "id": "deepseek-v3.2",
            "name": "DeepSeek V3.2 (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000002, "output": 0.0000004 },
            "contextWindow": 128000,
            "maxTokens": 8192
          },
          {
            "id": "glm-4.6v-flash",
            "name": "GLM Flash FREE (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0, "output": 0 },
            "contextWindow": 128000,
            "maxTokens": 4096
          },
          {
            "id": "qwen-max",
            "name": "Qwen Max (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000004, "output": 0.0000012 },
            "contextWindow": 32000,
            "maxTokens": 8192
          },
          {
            "id": "qwen-plus",
            "name": "Qwen Plus (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000002, "output": 0.0000006 },
            "contextWindow": 128000,
            "maxTokens": 8192
          },
          {
            "id": "qwen-turbo",
            "name": "Qwen Turbo (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.00000006, "output": 0.0000002 },
            "contextWindow": 128000,
            "maxTokens": 8192
          },
          {
            "id": "minimax-text-01",
            "name": "MiniMax 1M Context (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000002, "output": 0.0000016 },
            "contextWindow": 1000000,
            "maxTokens": 8192
          },
          {
            "id": "glm-4.6v",
            "name": "GLM 4.6V Vision (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000004, "output": 0.0000012 },
            "contextWindow": 128000,
            "maxTokens": 4096
          },
          {
            "id": "moonshot-128k",
            "name": "Moonshot 128K (NovAI)",
            "reasoning": false,
            "input": ["text"],
            "cost": { "input": 0.0000008, "output": 0.0000008 },
            "contextWindow": 128000,
            "maxTokens": 8192
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": { "primary": "novai/deepseek-v3.2" }
    }
  }
}

Choose Your Models

Here's every NovAI model and when to use it in OpenClaw:

Our Recommendation for OpenClaw Users

Set DeepSeek-v3.2 as your default model (already done in the config above). It gives you the best balance of coding quality and cost. Switch to other models when needed:

# In OpenClaw, switch models on the fly:
/model novai/deepseek-v3.2      # Best for coding (default)
/model novai/minimax-text-01    # Need to analyze a huge codebase
/model novai/qwen-turbo         # Quick tasks (commit msgs, comments)
/model novai/glm-4.6v-flash     # Free — testing or casual use

Verify & Start Coding

Let's make sure everything works:

# 1. Check that OpenClaw sees NovAI models
openclaw models list | grep novai

# Expected output:
# novai/deepseek-v3.2         DeepSeek V3.2 (NovAI)
# novai/glm-4.6v-flash        GLM Flash FREE (NovAI)
# novai/qwen-max              Qwen Max (NovAI)
# novai/qwen-plus             Qwen Plus (NovAI)
# novai/qwen-turbo            Qwen Turbo (NovAI)
# novai/minimax-text-01       MiniMax 1M Context (NovAI)
# novai/glm-4.6v              GLM 4.6V Vision (NovAI)
# novai/moonshot-128k         Moonshot 128K (NovAI)

# 2. Start OpenClaw — it will use DeepSeek by default
openclaw

# 3. Try a simple prompt to confirm it's working
> Write a Python function that reverses a linked list

If you get a response, you're done! OpenClaw is now running on NovAI.