Hugging Face integrates open source LLMs into GitHub Copilot Chat for VS Code.

Hugging Face has launched a significant integration that allows developers to connect Inference Providers directly with GitHub Copilot Chat in Visual Studio Code. This update facilitates access to and testing of open-source language models, such as Kimi K2, DeepSeek V3.1, GLM 4.5, among others, directly from the VS Code editor, eliminating the need to switch platforms or manage multiple tools.

A streamlined workflow

The process to initiate this integration is simple. Developers need to install the Hugging Face Copilot Chat extension, open the chat interface in VS Code, select the Hugging Face provider, enter their Hugging Face token, and then add the models they wish to use. Once connected, it is possible to seamlessly switch between providers and models using the model selection interface they are already familiar with.

However, a significant requirement has been pointed out in community discussions: the functionality requires having a recent version of the editor installed. AI researcher Aditya Wresniyandaka highlighted on LinkedIn that "the document forgot to mention that you need version 1.104.0 of VS Code, released in August 2025."

Expanding access to AI models

Historically, GitHub Copilot Chat has relied on a closed set of proprietary models. By linking with Hugging Face's network of Inference Providers, developers now have access to a much wider range of artificial intelligence tools, including experimental and highly specialized open-source models.

Muhammad Arshad Iqbal praised this initiative, stating: "Oh, this is great! Now we can use all those powerful open-source coding AIs directly in VS Code. No need to switch tabs just to try a model like Qwen3-Coder."

Advantages of the integration

This integration allows developers to use Copilot Chat with models optimized for specific programming tasks, industry sectors, or research domains, instead of being limited to default models. Moreover, the update ensures a service powered by Hugging Face's Inference Providers, granting developers access to hundreds of machine learning models through a single API.

The fundamental value proposition lies in unification: instead of managing multiple APIs with different levels of reliability, developers can query models from various providers through a consistent interface. Hugging Face highlights several benefits of this integration:

Instant access to cutting-edge models, surpassing what a single provider's catalog could offer.
No lock-in to a specific provider, as developers can switch between providers with minimal code changes.
Adequate production performance, with high availability and low latency in inference.
Developer-friendly integration, including direct compatibility with OpenAI's chat completion API and SDKs for Python and JavaScript.

Accessibility and pricing options

Hugging Face has designed the integration to be accessible. There is a free tier of monthly credits for testing purposes, while Pro, Team, and Enterprise plans offer additional capacity and pay-as-you-go pricing. According to Hugging Face, what developers pay is exactly the cost charged by the providers, with no additional markup.

Conclusions

The Hugging Face integration in GitHub Copilot Chat for VS Code represents a significant advancement in the access and application of open-source language models in software development. Developers can now leverage a broader variety of AI tools within their usual working environment, promising to enhance efficiency and innovation in their projects.

For more insights and articles on technology and software development, feel free to visit my blog.