Skip to main content

Use CodeGate with avante.nvim

avante.nvim is a Neovim plugin that provides a Cursor-like user experience. It integrates with multiple AI providers (e.g. Claude and OpenAI).

Install avante.nvim​

Install the avante.nvim plugin using your preferred plugin manager. For instructions, consult avante.nvim's documentation.

Configure CodeGate with avante.nvim​

Consult avante.nvim's documentation for specific examples on customizing your configuration using your preferred plugin manager.

Using lazy.nvim (recommended) add the following configuration to enable CodeGate with avante.nvim:

{
    "yetone/avante.nvim",
    -- ... etc ...
    opts = {
      -- your provider configuratio ...
      provider = "openai",
      openai = {
        endpoint = "http://localhost:8989/openai", -- use the endpoint matching your LLM provider
        model = "gpt-4o", -- your desired model (or use gpt-4o, etc.)
        timeout = 30000, -- timeout in milliseconds
        temperature = 0, -- adjust if needed
        max_tokens = 4096,
      },
    },
    -- ... etc ...
}
note

avante.nvim does not yet support fill-in-the-middle completions. You have to configure your fill-in-the-middle completion plugin separately.

Verifying configuration​

To verify the configuration, type :AvanteChat to start the avante.nvim plugin. You should see the avante.nvim UI appear in the Neovim window.

Type and send the following message:

codegate version

If CodeGate is configured correctly, you should see a response from the CodeGate server.

Model Muxing​

To take advantage of CodeGate's model muxing feature, use avante.nvim's OpenAI provider with the following configuration:

openai = {
  endpoint = "http://localhost:8989/v1/mux", -- use the endpoint matching your LLM provider
  model = "gpt-4o",
  timeout = 30000, -- timeout in milliseconds
  temperature = 0, -- adjust if needed
  max_tokens = 4096,
},