DevelopersIntegrationsCline

Cline with ARK

Add ARK as an OpenAI Compatible provider and run Cline's coding agent on open models with tool calling and a 256K-token context.

Cline: An open-source autonomous coding agent for VS Code, also available for JetBrains IDEs and the command line.

Before you start

You need an ARK API key: sign in to ARK Cloud and copy it from the dashboard, or use a key from your own ARK deployment and its base URL. The setup below gives each of Cline's modes an ARK Cloud model: arklabsai/Qwen3.5-122B-A10B plans, arklabsai/Qwen3.6-35B-A3B acts. Both call tools and read up to 256K tokens.

1. Install

Install
code --install-extension saoudrizwan.claude-dev

2. Connect it to ARK

Open Cline's settings (the gear icon) and turn on Use different models for Plan and Act. Then choose OpenAI Compatible as the API provider and fill in the fields for each mode:

Cline settings
Use different models for Plan and Act   on

Plan mode
  API Provider        OpenAI Compatible
  Base URL            https://api.ark-labs.cloud/api/v1
  API Key             your-ark-api-key
  Model ID            arklabsai/Qwen3.5-122B-A10B
  Context Window      262144
  Max Output Tokens   8192

Act mode
  API Provider        OpenAI Compatible
  Base URL            https://api.ark-labs.cloud/api/v1
  API Key             your-ark-api-key
  Model ID            arklabsai/Qwen3.6-35B-A3B
  Context Window      262144
  Max Output Tokens   8192

Context Window and Max Output Tokens are under Model Configuration. Save the settings and start a task in Cline's chat: switching between Plan and Act switches the model with the mode.

Good to know

  • Cline uses the Base URL exactly as entered, so include /api/v1.
  • Enter the context window yourself under Model Configuration; 262144 is the limit for the large Qwen models on ARK.
  • Code completion while you type needs a fill-in-the-middle endpoint, which ARK does not serve: use ARK for chat and agent work.
  • Thinking works with tools: when the agent turns thinking on, the Qwen models think before they call a tool (see Reasoning).
  • These tools send the whole conversation on every turn. ARK's prompt cache speeds up the repeated part of each prompt; every prompt token is still billed.

Next

Read about tool calling and limits and capacity on ARK, or browse other integrations.

Configuration keys follow Cline's official documentation as of November 2026. If the tool has changed since, its docs take precedence.