Suanlimao
← Back to all tutorials

Connect your tools

JetBrains AI Assistant

Add an OpenAI-compatible provider to AI Assistant in IntelliJ, PyCharm and other JetBrains IDEs.

About 6 minFor:Developers who work in JetBrains IDEs

Before you start

  • A JetBrains IDE with the AI Assistant plugin (2026 releases)
  • A key and a model name copied from the model plaza

Step 1 · STEP 0x01

Add an OpenAI-compatible provider

Settings | Tools | AI Assistant | Providers & API keys
1

Provider

OpenAI-compatible
2

URL

https://api.suanlimao.shop/v1
3

API Key

<YOUR_API_KEY>
4

Tool calling

On (if the model supports it)
5
Test Connection
Diagram only — the real screen in your version may differ

Open Settings | Tools | AI Assistant | Providers & API keys, choose OpenAI-compatible as the Provider under third-party providers, enter the address and key, and click Test Connection.

The official page does not say whether the URL needs /v1 (its only example is a local address). Try https://api.suanlimao.shop/v1 first; if Test Connection returns 404, use https://api.suanlimao.shop.

Create a key →

Step 2 · STEP 0x02

Assign features to the model

AI Assistant → Model Assignment
1

Core features

<MODEL_NAME> (your provider)
2

Instant helpers

<MODEL_NAME> or a cheaper model
Diagram only — the real screen in your version may differ
  • After connecting, under Model Assignment point Core features (chat, generation) and Instant helpers (small inline helpers) to your model; they can differ, and a cheaper model suits Instant helpers.
  • The docs do not say whether models are fetched or typed; if the UI asks you to enter one, use the name copied from the plaza.
Don't guess the model name: open the model plaza, click “Copy model id” on a model card and paste it. A wrong model name is the most common error.

Open the model plaza →

Step 3 · STEP 0x03

Code completion is a separate setting (optional)

AI Assistant → AI Completion
1

Provider

OpenAI Compatible
2

Base URL

https://api.suanlimao.shop/v1
3

API key

<YOUR_API_KEY>
4

Model

<MODEL_NAME>
5

Model context / Max output tokens

Per the model's details
Diagram only — the real screen in your version may differ
  • Inline completion does not use Model Assignment; set OpenAI Compatible with address, key and model separately in the AI Completion section.
  • Official note: Test Connection in this section does not verify the completion model.
Per the docs, “some features may be unavailable” with third-party models; your own models work without a JetBrains AI subscription, but a subscription is still needed for every feature.

✓ How to tell it worked

  • AI Assistant chat answers with your model.
  • The Usage page shows the request.

FAQ

Test Connection fails.

Check the key is complete, try the URL with and without /v1, then see the connection troubleshooting tutorial.

Chat works but some features fail or are greyed out.

That is the documented “some features may be unavailable”. Try turning on Tool calling (the model must support it) or assign a different model to those features.

Related tutorials

Still stuck?

Read the troubleshooting tutorials first. If that does not help, open a ticket with the request_id, time, model and the exact error (never paste your key).