Mike Croucher is a Customer Success Engineer for Research and Education who previously spent 20 years working in academic research computing. He writes about MATLAB, MathWorks, and intersections with other technologies for those who are changing the world.
Everyone's talking about Large Language Models (LLMs) and a huge number of you are using them too. Here are 4 ways to make use of them in the MathWorks ecosystem right now, no matter what your skill level is.
1. MATLAB AI Chat Playground
First released in 2023 with Chat GPT 3.5 Turbo as the underlying model, the MATLAB AI Chat Playground has recently been upgraded to use GPT-4o mini. As well as talking to the AI in the left hand side of the browser window, you can also run MATLAB code in the right hand-side using an engine that is based on the latest release of MATLAB, R2024b. AI and MATLAB together without needing to leave your web browser. This is probably the easiest way for MATLAB users to get started with experimenting with AI models.
2. Large Language Models (LLMS) with MATLAB
As a programmer, I have more fun with LLMS when I can interact with them programmatically. That's where this MathWorks repository comes in. It contains code to connect MATLAB to theOpenAI® Chat Completions API(which powers ChatGPT™), OpenAI Images API (which powers DALL·E™),Azure® OpenAI Service, and both local and nonlocalOllama™models. This allows you to leverage the natural language processing capabilities of large language models directly within your MATLAB environment.
For a great example of what's possible with this repository, see Toshi Takeuchi's personal project MatGPT which allows you to easily access OpenAI's API with an app.
3. GitHub Co-pilot with the MATLAB extension for Visual Studio Code
This one is for users of the ChatGPT site. The custom MATLAB GPT currently uses the GPT-4o model but has some additional customizations especially for MATLAB users. In particular, the AI uses Retrieval Augmented Generation (RAG) to search the MathWorks website before formulating an answer
댓글
댓글을 남기려면 링크 를 클릭하여 MathWorks 계정에 로그인하거나 계정을 새로 만드십시오.