Skip to main content

Autocomplete

The Autocomplete feature empowers you to choose from a curated selection of models, including those from Mistral and Ollama. This advanced tool enhances your coding experience by providing accurate and contextually relevant code suggestions.

How to Use:​

  • Go to Menu on CodeGPT sidebar:
    • Click on the Autocomplete option
    • Check the Status
    • Choose the AI model
    • Provider:
      • CodeGPT Plus
      • CodeGPT Plus-turbo
      • Mistral-codestral-latest
      • Ollama-qwen2.5-code:0.5b
      • Ollama-qwen2.5-code:1.5b
      • Ollama-qwen2.5-code:3b
      • Ollama qwen2.5-code:7b
      • Ollama-qwen2.5-code:14b
      • Ollama-qwen2.5-code:32b
      • Ollama deepseeek-coder:base
      • Ollama deepseek-coder-v2
      • Ollama codellama:code
      • Ollama codegemma:code
    • ⚠️IMPORTANT⚠️: Ensure that Ollama is running locally and that the model you are using is properly installed. For Mistral Codestral, insert the API Key and set Mistral as the main provider.
    • By default, the AI model determines the maximum number of tokens. Please refer to each provider's documentation for specific details. Additionally, the suggestion is generated with a delay of 3000 milliseconds.
Autocomplete settings

  • Effortless Integration: Seamlessly incorporate the suggested code snippets into your project, improving code quality and efficiency. Press Tab to accept the suggestion.
Autocomplete