Hi HN, we built Monty: a VS Code extension that connects your code with an LLM.
We designed Monty to be privacy friendly: your chat messages are sent directly to your LLM (bring your own key) and we use a local index for RAG (as opposed to a remote service for embedding/storage).
If your company restricts tools that send code to 3rd party servers then maybe Monty is for you.
Monty supports multiple LLM providers: OpenAI and OpenAI-compatible providers (including Groq and Ollama), Azure OpenAI, Anthropic, and Google AI Studio.
Monty has a chat-style interface with 2 main context options: - (default) Chat using user-specified code context - Chat using automatic code context (RAG)
Monty doesn’t restrict token usage. For example, if you are using gpt-4-turbo with a 128k token limit you are free to utilize all of it. Load huge files or dozens of functions/classes into the code context and ask whatever you want.
You can use Monty for free if you bring your own API key. Without an API key new accounts get 40 chat messages on us.
We’d love to get your feedback. For more information check out our website https://www.ai-maintainer.com/ or post a question here.
