wylon

Configuring Openclaw

Openclaw is a coding agent available as a CLI, a TUI, and a web dashboard, connecting to models over the OpenAI-compatible protocol.

1. Install

Linux / macOS

curl -fsSL https://openclaw.ai/install.sh | bash

Windows PowerShell

iwr -useb https://openclaw.ai/install.ps1 | iex

2. Run the setup wizard

openclaw onboard --install-daemon

Follow the prompts in order:

  1. Personal use: select Yes;
  2. Setup mode: choose QuickStart;
  3. Model provider: choose the Custom / OpenAI-compatible option;
  4. Base URL: enter https://api.wylon.cn/v1;
  5. API Key: paste your wylon API Key;
  6. Model name: enter the model ID you want, e.g. moonshotai/Kimi-K2.6 or deepseek-ai/DeepSeek-V4-Flash;
  7. Remaining options (messaging channels, Skills, etc.) can be configured as needed, or skip with Skip for now.
tips_and_updates
Openclaw ships with several built-in vendor presets, but wylon is not one of them — at the "Model provider" step, choose "Custom / OpenAI-compatible" and enter the details manually.

3. Use

openclaw dashboard   # open the web console
openclaw tui         # launch the terminal UI
openclaw terminal    # terminal chat mode
沪ICP备2026010432号-1 沪公网安备31010402336632号