Continue Blog

Use QwenCoder2.5 32B on the SambaNova Cloud and never wait for replies

Use QwenCoder2.5 32B on the SambaNova Cloud and never wait for replies

We are excited to partner with SambaNova Systems to launch QwenCoder2.5 32B on their cloud! In October, they published a blog post about how to use Continue with Llama 3.1 405B. Today, Continue developers can now experience the power of getting suggestions from QwenCoder2.5 32B on the

Model Context Protocol x Continue

Model Context Protocol x Continue

Last week, Anthropic introduced the Model Context Protocol (MCP), a new open protocol that "enables seamless integration between LLM applications and external data sources and tools". Also last week, Continue added support for MCP! Let’s explore why MCP is well-suited for Continue and how you can use

Instant Apply

“UX risk” Over the last year and a half of developing features that users didn’t like, finding out they didn’t like them, and then trying again, we’ve built a better idea of what will and won’t fly. Among the biggest predictors is what I call “UX

A custom autocomplete model in 30 minutes using Unsloth (Community post)

A custom autocomplete model in 30 minutes using Unsloth (Community post)

This is a guest post on the Continue Blog by Sophia Parafina, a developer advocate who has previously worked at Pulumi, Anaconda, and Docker. Continue is an open-source AI code assistant. It's a compound AI system that uses a set of models (chat, autocomplete, embeddings, etc) along with

Protecting the Continue community

Protecting the Continue community

Continue is the leading open-source AI code assistant, already having tens of thousands of users, nearly 200 contributors, and a vibrant, growing community. We envision an ecosystem of products that build with and around Continue, and that is part of why we released our project under an Apache 2.0

Root path context: The secret ingredient in Continue's autocomplete prompt

Root path context: The secret ingredient in Continue's autocomplete prompt

If you interpreted “root path” as the root directory of your computer, don’t worry! Continue wants nothing to do with this. What we’re going to share is our strategy for gathering (and caching) autocomplete context, which relies heavily on the root path: the path from a node in

Why we’re excited about prompt caching for engineering teams

Why we’re excited about prompt caching for engineering teams

Anthropic’s rollout of prompt caching, also known as context caching, follows similar offerings from providers like DeepSeek and Google and has been getting a lot of attention recently. At Continue, we're particularly excited about the potential impact on engineering teams, as prompt caching can significantly reduce costs

Happy 1st birthday, Continue!

Happy 1st birthday, Continue!

Looking back on the first year of Continue Just over a year ago, we launched Continue on Hacker News. Since then, the Continue community has taken shape and grown considerably: * 300k+ downloads across VS Code and JetBrains extensions * 14k+ stars on GitHub, 5k+ Discord members, and 140+ contributors * Lots of

A framework for evaluating AI code assistants

A framework for evaluating AI code assistants

Throughout hundreds of conversations with users, we’ve heard lots about what they need in a great AI code assistant. In this post we’ll share a set of criteria that are representative of those we find companies caring about the most, especially once they have the hindsight of running

An entirely open-source AI code assistant inside your editor

An entirely open-source AI code assistant inside your editor

This was originally a guest post on the Ollama blog that covers how to set up, explore, and figure out the best way to use Continue and Ollama together. Continue enables you to easily create your own coding assistant directly inside Visual Studio Code and JetBrains with open-source LLMs. All

Codestral x Continue

Codestral x Continue

We are excited to share that we have partnered with Mistral AI to launch their latest model, Codestral! 0:00 /0:19 1× Codestral was trained on more than 80 programming languages and is tailored for coding tasks like fill-in-the-middle and code completion. In fact, our early testing of Codestral

How to measure the impact of AI software development systems

How to measure the impact of AI software development systems

Empowering your developers with an AI software development system, which they can use to automate more tasks in their job, will improve engineering outcomes and quality of life for developers at your organization. For example, ML-enhanced code completion was shown to improve developer productivity at Google and GitHub. Below we

How can you improve the code suggestions you get from LLMs?

How can you improve the code suggestions you get from LLMs?

Josh Collinsworth wrote a post titled “I worry our Copilot is leaving some passengers behind” a couple weeks ago, and I can’t stop thinking about it, especially this section in particular: "Why should we just accept that LLM tools not only fail to at least give us the