Setting Up a Local AI Co-Pilot for PyCharm with Continue and Ollama

local_ai_copilot_pycharm.md

Steps:

1. Install the Continue plugin:

   PyCharm --> Settings --> Plugins --> search for "Continue" --> Install

2. Download and install Ollama

3. Download some LLM model, which can be run locally. For example, the open-source Granite model from IBM:

ollama pull granite3.1-dense:8b

4. OPTIONAL: In addition to the LLM for chat and code generation, you can install an embedding model to enable the Retrieval Augmented Generation (RAG) capabilities of Continue, for example:

ollama pull granite-embedding:30m

5. Configure the Continue PyCharm plugin

Modify ~/.continue/config.json file, for example:

{
  "models": [
    {
      "title": "Granite 3.1 8b",
      "provider": "ollama",
      "model": "granite3.1-dense:8b"
    }
  ],
  "tabAutocompleteModel": {
    "title": "Granite 3.1 2b",
    "provider": "ollama",
    "model": "granite3.1-dense:2b"
  },
  "embeddingsProvider": {
      "provider": "ollama",
      "model": "granite-embedding:30m",
      "maxChunkSize": 512
  },
  "customCommands": [
    {
      "name": "list-comprehension",
      "prompt": "{{{ input }}}\n\nRefactor the selected python code to use list comprehensions wherever possible. Present the output as a python code snippet.",
      "description": "Refactor to use list comprehensions"
    }
  ]
  ...
}

Sources:

• IBM: Build a local AI co-pilot using IBM Granite Code, Ollama, and Continue