JetBrains AI Assistant with ARK
Add ARK as an OpenAI-compatible provider and use open models in AI Chat and JetBrains' agents, with tools from your MCP servers.
JetBrains AI Assistant: The AI chat, code generation and agent host in JetBrains IDEs such as IntelliJ IDEA, PyCharm and WebStorm.
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 uses three ARK Cloud models: arklabsai/Qwen3.5-122B-A10B for AI Assistant's core features, arklabsai/Qwen3.5-4B for its instant helpers, and arklabsai/Qwen3.6-35B-A3B to pick in chat. All three call tools; the 122B and the 35B read up to 256K tokens, the 4B up to 128K.
1. Install
Use a JetBrains IDE with AI Assistant 2026.2 or later. Models from your own provider work without a JetBrains AI subscription, though some features may be unavailable with them.
2. Connect it to ARK
Open Settings | Tools | AI Assistant | Providers & API keys, add an OpenAI-compatible provider under Third-party AI providers and fill in:
Provider OpenAI-compatible
URL https://api.ark-labs.cloud/api/v1
API Key your-ark-api-key
Tool calling onClick Test Connection, then Apply. AI Assistant does not assign models from an OpenAI-compatible endpoint to its features by itself: until you do, they are available only in AI Chat. Assign them under Models Assignment:
Core features arklabsai/Qwen3.5-122B-A10B
Instant helpers arklabsai/Qwen3.5-4BCore features covers in-editor code generation and commit messages, and is the default model in AI Chat. Instant helpers covers lightweight features such as chat context collection, chat title generation and name suggestions. Pick arklabsai/Qwen3.6-35B-A3B, or any ARK model, in the model selector of AI Chat and of compatible agents.
Good to know
- Turn on Tool calling so ARK models can call tools from your MCP servers.
- Organizations managed through JetBrains IDE Services may turn third-party providers off.
- 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 JetBrains AI Assistant's official documentation as of November 2026. If the tool has changed since, its docs take precedence.