Qt AI Assistant
Boost Productivity with AI: Focus on Coding, Not Chores
Qt AI Assistant helps developers focus on what they love doing the most: programming. By offloading developers from tasks such as writing test cases, code documentation, explanations, and boilerplate QML code, it frees time for coding.
Best-in-Class QML
While open to connecting to your favorite LLMs, Qt AI Assistant delivers best-in-class QML with our fine-tuned Llama for models, trained on over 4000 QML applications created and validated by experienced QML programmers.
Test Cases and Documentation
Optimized to work on Qt Test syntax, Qt AI Assistant readily delivers comprehensive unit tests and clear documentation for any portions of your code.
Next-Gen Expertise
Be it QML, C++, or Python, by selecting a portion of code, Qt AI Assistant helps you explain, refactor, and optimize code, as well as find the latest and most advanced features that can take your application to the next level.
AI-Powered Productivity, Human-Powered Creativity
Optional Auto-Completion
Qt AI Assistant can automatically complete code as you type, prompting suggestions that can be accepted all-in-one, line by line, or word by word. If you prefer not to consume LLM tokens or be less disrupted by Gen AI, you can trigger code completion on request by using keyboard shortcuts or prompts.
Open Access to Your Favorite LLMs
By default, Qt AI Assistant is configured to connect to LLMs of Meta’s Llama series, Anthropic's Claude 3.5 Sonnet, or OpenAI GPT4o. However, you are free to connect to any commercial cloud services or self-hosted LLMs. You can also leverage different LLMs for different use cases: if a general-purpose model is not enough, you can pick more specialized ones based on the relevant use case.
Unit Testing
You can quickly improve your code testing coverage by creating unit test cases with Qt AI Assistant. Simply select the QML component, method, or class you need to test, and you'll get multiple test cases ready for the Qt Test project.
Code Documentation
Improve code maintainability and collaboration, help developers understand the purpose, structure, and functionality of your code, by asking Qt AI Assistant to generate the documentation for any portion of your code.
For Developers
Focus on creating unique, engaging user experiences for cross-platform applications while generative AI takes care of boilerplate UI code, test cases, and documentation. Qt AI Assistant offers you assistance on writing QML code, with examples and explanations.
For R&D Managers
Access the best LLM for each task. While delivering the best QML/Qt Quick coding performance out-of-the-box, Qt AI Assistant can also smartly route user requests based on the task and programming language to access more specialized or cost-efficient LLM.
For Product Managers
Maintain and expand your competitive edge by automating software creation on secure LLMs. Qt AI Assistant ensures your code is not leaked accidentally or through LLM training by running your own LLM on-premise or on a cloud of your choice.
Learn More
Read additional resources to learn more about Qt AI Assistant.
AI Assistant Documentation
Get a more detailed view into how you can utilize Qt AI Assistant in your projects.
The Making of Qt AI Assistant
This blog post gives behind-the-scenes view of how Qt Assistant came to life.
Frequently Asked Questions
Do I need a commercial license to use the Qt AI Assistant?
- AI Assistant
Why can't I use the Qt AI Assistant with a Qt for Application Development Professional license?
- AI Assistant
Do I need a Large Language Model to use the Qt AI Assistant?
- AI Assistant
Where can I get a Large Language Model to use for the Qt AI Assistant?
- AI Assistant
There are three ways to connect a Large Language Model to the Qt AI Assistant:
- Subscribe to a Large Language Model cloud service such as Anthropic or OpenAI.
- Run a Large Language Model such as Llama 3.3 70B and Code Llama 13B in your private cloud.
- Run a Language Model such as CodeLlama 7B-hf or CodeLlama 13B-hf on your computer locally using the Ollama technology.
If you need help running LLMs in your private cloud, then we can suggest one of our professional service partners.
Is Qt using any of my data to train a Large Language Model?
- AI Assistant
No, the Qt Company doesn't use any input or outputs of LLMs to train any model unless you explicitly ask the Qt Company to do so in writing. The Qt Company is not part whatsoever in the actual generation of outputs. The Qt AI Assistant in Qt Creator communicates directly with the LLM.
The legal way to say this would be:
"The Outputs generated for Customer will be considered the Customer’s Outputs, and The Qt Company will not claim any right to, title to, or interest in Customer’s Outputs. Notwithstanding the foregoing, any The Qt Company Intellectual Property contained within the Outputs remains the intellectual property of The Qt Company and its licensors and is subject to the applicable terms and conditions regarding such intellectual property. Customer acknowledges that Outputs are generated as a non-exclusive response to the work with Qt AI, so the same or similar Outputs, or other suggestions may also be generated for other Qt AI customers, based on inputs or context of use. In regards to Qt AI Assistant, the Outputs are generated by the use of Qt AI Assistant in conjunction with Third-Party Software and as such, may be subject to third-party rights, including open-source licenses. Consequently, Customer must ensure that the Output generated by Qt AI Assistant does not violate third-party rights, such as open-source software licenses."
Can the Qt AI Assistant create code for Qt Widgets or code compatible with Qt 5 releases?
- AI Assistant
How can I try out the Qt AI Assistant?
- AI Assistant
Trying out Qt AI Assistant is straightforward:
- Install or upgrade Qt Creator to the latest release Qt Creator 15.0.1 or newer (without it Qt AI Assistant won't work)
- In Qt Creator, go to the Extensions view and select the AI Assistant
- Install the Qt AI Assistant plug-in (the thing that plugs in to your IDE)
- Accept Internet access and the installation of the AI Assistant Server (the component that talks to the LLMs in the background)
- Select installation scope (individual user / all users) and enable that extension is loaded at start-up
- Connect to at least one Large Language Model
Note: You need to have a valid premium Qt Development license (Qt for AD Enterprise or better). If you don't have one, sign up for a Qt Development evaluation license.