Local Llm Code Completion

via “multi-model-aware tab autocomplete with lsp context integration”

Open-source AI code assistant for VS Code/JetBrains — customizable models, context providers, and slash commands.

Unique: Decouples autocomplete model selection from chat/edit models via a unified LLM abstraction layer that supports 40+ providers, and integrates LSP context directly into the completion pipeline rather than relying on simple token-based context windows. The Next Edit feature uses IDE-aware cursor tracking to predict multi-line completions.

vs others: Unlike Copilot (locked to OpenAI) or Cursor (limited provider choice), Continue allows independent model selection per feature and works with local models, reducing latency and API costs for teams with data sovereignty needs.