Can't get local deepseek-coder-v2:16b model working with Avante

[careb0t]

        provider = "deepseek";
        vendors = {
          deepseek = {
            __inherited_from = "openai";
            endpoint = "http://localhost:11434";
            model = "deepseek-coder-v2";
          };
        };

This is my configuration for Avante (in nix as I am using NixOS) trying to use the local Deepseek model that I have running with Ollama. When I go to http://localhost:11434 in my browser, it shows me that Ollama is running, and I can see the Ollama service when I check btop. When I open up the Avante ask UI panel and enter in a question, I get a 404 error.

And here is the output of ollama list showing the model is installed:

Has anyone had success getting a locally running Deepseek coder model via Ollama to work properly in Avante? I feel like I must be close, but am just missing something I don't see. I have been able to get Avante to use online models with an API endpoint, but not a local model yet.

[jolo-dev]

I believe you are missing

api_key_name = "" endpoint = "http://127.0.0.1:11434/v1"

So it should be

provider = "ollama";
        vendors = {
          deepseek = {
            __inherited_from = "openai",
           api_key_name = "",
            endpoint = "http://localhost:11434/v1",
            model = "deepseek-coder-v2",
          };
        };