Skip to main content
Qwen Cloud home page
Docs
Clients & tools

OpenClaw

Open-source AI coding assistant

OpenClaw is an open-source AI assistant that helps you write, debug, and refactor code. Connect it to Qwen Cloud's pay-as-you-go API for powerful AI assistance through the OpenAI-compatible endpoint.

Quick start

Get running in a few minutes: 
# 1. Install
npm install -g openclaw@latest
openclaw --version

# 2. Configure (run onboard to start the setup wizard)
openclaw onboard

# Follow the interactive prompts:
# "I understand this is powerful..." -> Yes
# Onboarding mode -> QuickStart
# Model/auth provider -> Skip for now (configure Qwen Cloud later)
# Filter models by provider -> All providers
# Default model -> Keep current
# Select channel -> Skip for now
# Configure skills now? -> No
# Enable hooks? -> Space to select, Enter to continue
# How do you want to hatch your bot? -> Do this later

# Configure Qwen Cloud API
# Base URL: https://dashscope-intl.aliyuncs.com/compatible-mode/v1
# API Key: sk-xxx
# Model: qwen3.5-plus

# 3. Test
openclaw gateway
openclaw tui
# Type: "Write a Python function to sort a list"
# You should see: OpenClaw generates the sorting function

Configuration

Installation options

  • Quick install (npm)
  • Script install
npm install -g openclaw@latest
openclaw --version

Complete initial setup wizard

After installation, OpenClaw automatically starts the setup wizard. You can also run openclaw onboard manually.
Configuration itemRecommended setting
"I understand this is powerful..."Yes
Onboarding modeQuickStart
Model/auth providerSkip for now (configure Qwen Cloud later)
Filter models by providerAll providers
Default modelKeep current
Select channelSkip for now
Configure skills now?No
Enable hooks?Press Space to select, Enter to continue
How do you want to hatch your bot?Do this later

Configuration methods

  • Interactive setup
  • Config file
openclaw dashboard
In the dashboard, go to Config > RAW and replace with Qwen Cloud API information.
"models": {
  "mode": "merge",
  "providers": {
    "qwencloud": {
      "baseUrl": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
      "apiKey": "DASHSCOPE_API_KEY",
      "api": "openai-completions",
      "models": [
        {
          "id": "qwen3.5-plus",
          "name": "qwen3.5-plus",
          "reasoning": false,
          "input": ["text", "image"],
          "contextWindow": 1000000,
          "maxTokens": 65536
        }
      ]
    }
  }
},
"agents": {
  "defaults": {
    "model": {
      "primary": "qwencloud/qwen3.5-plus"
    },
    "models": {
      "qwencloud/qwen3.5-plus": {}
    }
  }
}
Free quota and billing:
  • First-time users get a free quota (valid for 90 days)
  • Enable Free quota only to prevent unexpected charges
For Coding Plan users: Use your exclusive base URL and API key instead. See Coding Plan for OpenClaw.

Model recommendations

TaskModelWhy
Quick tasksqwen3-coder-nextFast, cost-effective
General codingqwen3.5-plusBalanced performance
Complex logicqwen3-coder-plusDeep understanding
Architectureqwen3-maxAdvanced reasoning

Limitations

  • Node.js 22+: Requires Node.js version 22 or later
  • Model compatibility: Not all Qwen models support all features

Examples

Interactive chat

openclaw gateway  # Start the gateway
openclaw tui      # Open terminal UI
# Type your questions directly

Model switching

Use the provider prefix to ensure the correct API endpoint: 
openclaw models set qwencloud/qwen3-coder-plus

Advanced features

Dashboard

Access the web dashboard: 
openclaw dashboard

Custom agents

Create specialized agents for different tasks: 
openclaw agents add
# Follow the prompts to create a new agent

Troubleshooting

"Invalid API key" error
Solution:
  • Verify API key is correct
  • Ensure API key has quota
"Model not available" error
Solution:
  • Check model ID spelling
  • Verify model access in your workspace
  • See Model list
High token consumption
Solution:
  • Use focused prompts
  • Choose appropriate models for tasks

Token optimization

Reduce usage by:
  1. Specific prompts: Be precise about what you need
  2. Model selection: Use lightweight models for simple tasks
  3. Directory focus: Work in specific project folders
  4. Batch operations: Combine related tasks in one session
OpenClaw | Qwen Cloud