This extension does not come with a built in backend for running LLMs. Instead you are able to use any existing tool that supports the OpenAI API format. Like the Oobabooga WebUI and many other
Features
Inline (multi line) code completion
Works with any OpenAI compatible API
Save multiple API Endpoints and switch easily between them
Reducing requests to LLMs by
saving previous responses
skiping completion depending on the last symbol
only posting request if no input was given for some time (can be specified in the settings)
Dynamically detect multi line or single line completion
Add other files to the completion context to improve the output
Roadmap
Improve (optional) features to reduce LLM requests
Add option (possible regex) to specify after which characters the LLM should be/not be triggered
Increase context
Add content after cursor to prompt
Workspace specific Context selection view
Add token count
Add clear all button
Check directory checkbox if all files are selected
Project Context
Add complete project to context using embeddings or similar
Return multiple completions (add suggestions from history)
Improve detection of already existing symbols at the end of a completion
Reduce chance of repeating already existing symbols
Reduce completion stopping because of false detection of already existing symbols
Custom completion stopping
Detect bracket/brace/parenthesis imbalance and stop/don't stop
Missing closing bracket -> don't stop
Improve detection of already existing symbols at the end of a completion based on this
First line preview
Preview the first line (and maybe more) of the completion while completion is still ongoing
Enable/Disable llm completion for specific filetypes
Extension Settings
localcompletion.active_endpoint: The URL of the API which is used for generating the code completion
localcompletion.endpoints: List of URL endpoints
localcompletion.temperature: Temperature of the LLM
localcompletion.max_tokens: Maximum number of tokens in the response
localcompletion.reduce_calls: Reduce API calls with various strategies (e.g. skip completion if last symbol was a letter)
localcompletion.skip_autocomplete_widget: Skip completion if autocomplete widget is active
localcompletion.completion_timeout: Minimum time between keystrokes (in ms) before sending a completion request (Reduces API calls, which are closed immediately after)
localcompletion.max_lines: Maximum number of lines in the response (empty lines are ignored)
localcompletion.add_visible_files: Add all visible files to completion context
localcompletion.context_files: List of files to add to completion context (should usually not be edited manually)
localcompletion.context_gitignore: Whether to ignore files in the .gitignore in the context selection view
Known Issues
OpenAPI keys
The extension does not yet support a custom API key. This means it only works for APIs which do not need a key.
Model switching
Model switching is not supported at the moment as most local tools don't support that property either.
Context selection
Symlinks can cause problems with additional context selection. They are not handled properly at the moment.
Selected files in the .gitignore are not automatically removed from the additional when "Apply .gitignore to context" is checked
No git installed
In order to automatically ignore files in the .gitignore for the context I use a package which interacts with git. At the moment, I was not able to test the extension without git installed. If you encounter any issues please let me know.
