Skip to content

Conversation

@penso
Copy link
Owner

@penso penso commented May 23, 2025


NOTE
Please verify that the base repository above has the intended destination!
Github by default opens Pull Requests against the parent of a forked repository.
If this is your personal fork and you didn't intend to open a PR for contribution
to the original project then adjust the base repository accordingly.


google-labs-jules bot and others added 11 commits May 23, 2025 21:36
This commit updates the Minuet plugin configuration to use an LM Studio
backend, instead of the previous Ollama setup.

- Modified `lua/custom/plugins/minuet-ai.lua`:
  - Switched the active provider to 'openai_compatible'.
  - Ensured the endpoint and model name point to your
    LM Studio instance (`http://m4max.local:1234/v1/chat/completions`
    with model `qwen2.5-coder-3b-instruct`).
  - Commented out the previous 'openai_fim_compatible' (Ollama) provider.

- Verified `init.lua`:
  - Confirmed that `blink.cmp` is already correctly configured to use
    Minuet as a completion source via `minuet.blink`. No changes
    were needed in this file.

I have instructed you on how to test this new configuration by
ensuring LM Studio is running with the specified model and server settings,
restarting Neovim, and attempting to trigger AI completions via `blink.cmp`.
Temporarily switches Minuet plugin to use the official OpenAI API
for debugging purposes.

- Modifies `lua/custom/plugins/minuet-ai.lua`:
  - Sets provider to 'openai'.
  - Configures `api_key` to be read from the `OPENAI_API_KEY`
    environment variable.
  - Sets OpenAI endpoint to `https://api.openai.com/v1/chat/completions`
    and model to `gpt-3.5-turbo`.
  - Enables debug notifications (`notify = 'debug'`).
  - Comments out the previous LM Studio (`openai_compatible`) configuration.

This change is intended to isolate and test Minuet's core functionality
before further debugging of LM Studio integration.
Comments out the explicit DAP configuration within `rustaceanvim`'s
setup. This is to rely on `mason-nvim-dap` to provide the global
configuration for `codelldb`, which `rustaceanvim` should then use.

This may resolve an error related to `dap.adapter.executable.command`.
@penso penso merged commit 78c8d54 into master May 24, 2025
1 check passed
@penso penso deleted the feat/enhance-rust-nvim-config branch May 24, 2025 12:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants