Would you mind adding a sample neovim plugin for this too?

[afidegnum]

As I'm exploring this, I will be glad if this can be integrated into neovim as well.
Please do give a provision for additional model (a local one like ollama especially)

[donhardman]

Hi, it would be great to hear from you how you see it being done in Neovim as a plugin. In terms of UI and interaction, if you have good references, I would be happy to see them. This is the way we think about providing the best and still simple experience for integration.

As for Llama, I will check it out. You run it locally or through some provider?

[afidegnum]

I've been experimenting with the CLI and will highlight some needed functionalities. For Ollama, you can download and run models locally, provided you have a compatible GPU or sufficient CPU and RAM. Popular coding models include Qwen2.5, CodeLlama, DeepSeek, and others. Running these locally is cost-effective compared to using services like OpenRouter.