Create, Edit, Analyze: ChatGPT Code Interpreter
Create, Edit, Analyze: ChatGPT Code Interpreter
Interactive AI conversations for any level of experience
Total reading time is about 3 minutes.
Advertise your AI job, event, or tool with Visually AI’s community of over 5K subscribers:
🔮AI News
Spotlight: ChatGPT Code Interpreter
The world of AI has been buzzing with the recent release of the ChatGPT Code Interpreter.
The tool is available for ChatGPT Plus users and must be enabled in the settings.
It lets you run code and access files you've uploaded. This means you can request ChatGPT to perform a variety of tasks, such as data analysis, chart creation, file editing, and even complex mathematical computations.
I tried Code Interpreter a few times this week, and I was most impressed by the step-by-step walkthroughs it shows you before it begins.
I don’t know Python, but it didn’t matter because it gave me options to accomplish my request to animate images, explained each step while it worked, and detailed workarounds for issues:
It gives you a link to download the file to your device, and you can ask for adjustments if you’re not satisfied with the results.
Here’s one of my first results from Code Interpreter animating four of my Midjourney-generated images:
It’s not a great animation tool, but the example above took less than 5 minutes!
I used it to generate scannable QR code links to my website and newsletter, several animated GIFs and videos using different uploaded images, and charts and graphs of my Twitter analytics from an uploaded CSV file.
I think it has a lot of potential for practical use, and it will be interesting to see creative ideas in the coming weeks.
Multimodal Bard
Google’s ChatGPT alternative, Bard got several updates this week, including:
Google Lens integration - upload images with prompts
Text-to-voice in 40+ languages to translate chat (includes Hindi, Arabic, German, & more)
Bard is multimodal, which means it can understand and work with different types of information like text, images, and audio all at once.
I asked Bard to explain three helpful ways we can use multimodal features with practical examples.
This is Bard’s answer with light editing by me:
Sure, here are 3 ways Bard can be used multimodal, with specific real-world examples:
Create interactive educational content that combines text, images, and audio. For example, a teacher could use Bard to create a virtual field trip to a museum, or a student could use Bard to create a narrated presentation about their favorite book.
Translate videos and live streams to make educational content more accessible to people who speak different languages or to provide live translation for business meetings or conferences.
Create multimodal customer support experiences that combine text, images, and video. This could be used to provide customers with self-service support options or to create more engaging and informative live chat experiences.
Some of the features included in Bard’s response might not be available at the moment, but it seems like there are plenty of ways we can use Bard and other large language models with these capabilities.
Anthropic’s Claude 2 Gets Major Updates
Anthropic’s ChatGPT competitor Claude 2 launched this week with major updates to its capabilities.
How does Claude 2 compare to ChatGPT?
Knowledge stops in 2023 vs. ChatGPT’s knowledge up to September 2021
Summarize up to 75,000 words vs. ChatGPT’s limit of 3,000 words
Claude 2 can answer questions similar to ChatGPT’s Code Interpreter with longer answers, detailed explanations, and conversations with users
Stability AI’s Stable Doodle
Stability AI's Stable Doodle is the latest sketch-to-image tool.
It’s available on ClipDrop, and it quickly turns doodles into decent images. You can try different styles to change the look and download the ones you like.
Look what it did with my flower sketch:
You can’t change the dimensions, and your images will be in a square, 1:1 format.
But you can customize the dimensions to portrait or landscape with ClipDrop’s Uncrop tool.
I used Uncrop’s generative fill abilities to turn my square image into one with portrait dimensions:
This adds to Stability AI’s recent launch of Stable Diffusion XL (SDXL 0.9) generative AI model, available on ClipDrop, DreamStudio, NightCafe, and Leonardo AI.
In partnership with Taskade
Experience the magic of Taskade AI right in your pocket
Generate tasks, brainstorm ideas, and chat with the AI assistant, now available on iOS and Android.
Enhance your workflows anywhere. Learn more.
🚀 This Week’s AI Tools
Verble: AI-powered speechwriting assistant empowers people with public speaking/pitching/verbal communication. Create an account with all PRO features unlocked for free until September 1st. (link)
Bullet Points: AI tool that briefs you on any subject in the news and will send you briefs as email newsletters. (link)
QuickQR Art: Generate scannable QR code art customized for your brand. (link)
🎁 Get it free: The AI Visual Creator’s Toolkit
Boost your content with my all-in-one, free visual AI toolkit!
Access AI-powered tools for AI-generated images, image editing, and more:
Thanks for reading and have a creative week!
Have you tried Claude 2 yet? I haven't. Very curious as to whether it matches up in the areas people are saying it does.