Shypd.aiThe largest open index of AI toolsllm-ls is an LSP server that leverages LLMs for code completion and other development enhancements. It provides a common platform for IDE extensions, handling LLM interactions to streamline the development experience.
Trending
About
llm-ls is an LSP (Language Server Protocol) server designed to integrate Large Language Models (LLMs) into the development workflow, primarily for code completion and other code-related enhancements. It acts as a foundational platform for building IDE extensions, abstracting away the complexities of interacting with various LLMs. The tool supports multiple backends, including Hugging Face's Inference API, text-generation-inference, ollama, and OpenAI compatible APIs like the Python llama.cpp server bindings. Key features include intelligent prompt generation using the current file as context, support for 'fill in the middle' completions, and tokenization to ensure prompts fit within the model's context window. It also parses the Abstract Syntax Tree (AST) to determine appropriate completion types (multi-line, single-line, or none). Additionally, llm-ls gathers telemetry for potential retraining, storing data locally without external export.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
