Continue
Open-source AI code assistant for VS Code and JetBrains that works with any LLM
About Continue
Continue was founded in June 2023 in San Francisco by Ty Dunn and Nate Sesti as part of Y Combinator's W23 batch. The project provides an open-source AI code assistant that integrates directly into VS Code and JetBrains IDEs, offering tab autocomplete, chat, inline editing, and context-aware code generation. The critical differentiator is model flexibility: unlike GitHub Copilot or Cursor which are tied to specific providers, Continue works with any LLM — OpenAI, Anthropic, local models via Ollama/LM Studio, or self-hosted open-source models. This makes it the leading option for organisations that want AI code assistance but need to control which models process their source code. Continue raised $5.1 million in total funding. Notable investors include Y Combinator, Heavybit, and angel investors including Julien Chaumond (co-founder of Hugging Face). For European development teams, Continue's open-source architecture offers maximum data sovereignty: organisations can run it entirely on their own infrastructure with locally hosted models, ensuring source code never leaves their network. This is particularly relevant for financial services, government, and defence contractors where source code is classified or highly sensitive. The Continue Hub allows teams to create and share custom AI assistants configured for specific codebases, coding standards, or domain knowledge.
TrustKit Score Breakdown
?68% StrongPricing
FreeFree tierQuick Facts
Frequently Asked Questions
Is Continue GDPR compliant?
Continue has a TrustKit compliance score of 68% (Strong). Data Residency: Fully self-hosted with local model inference. Source code never leaves the developer's machine when using local models. Maximum data sovereignty achievable.. Legal Jurisdiction: Delaware incorporation. However, the open-source licence and self-hosting capability mean the product can be used with zero dependency on the company's cloud infrastructure..
Where does Continue store data?
Continue hosts data in: Self-hosted with local models; no data sent externally if configured for local inference. Fully self-hosted with local model inference. Source code never leaves the developer's machine when using local models. Maximum data sovereignty achievable.
Does Continue train on user data?
Continue: Open-source: full control; no telemetry or data collection when self-hosted. Open-source, self-hosted, no telemetry when configured for local use. Complete data lifecycle control. Code is auditable.
What certifications does Continue hold?
No certifications have been confirmed for Continue yet. No formal certifications. Early-stage open-source project. Self-hosted model means organisations apply their own security controls.