Yeah, it is a bit hard to test w/o n access to bio feature without a VPN. I have old GPTs where the memory option was turned on when it was available but it seems like it has been removed completely
@testingcatalog
-
Testing Claude’s analysis of EU elections
By
–
Try Claude This is at least what it can say about EU elections
-
Proposed workflow for image processing using ChatGPT and messenger apps
By
–
Mobile app to take photos inside the ChatGPT app + run the code over there and then a messenger desktop app to receive the results back?
-
Speculation on upcoming AI features for Perplexity
By
–
Perplexity is cooking something Is it a new image gen model for making better banners for Pages? Is it a new vision capability? GPT-4o multimodal integration? Or @nothing
? …??…??.. -
User feedback on AI model selector functionality
By
–
It is a feature called “Model Selector”. But anything other than Sonnet isn't working anyway
-
Google’s Veo Model Faces Competition and Development Challenges
By
–
And Veo by Google feels already behind. Especially with its current “WIP” state
-
Comparison of generative AI video model quality
By
–
The length of videos and quality feels to be close to what Google has shown with VideoFX
-
Using LLM context windows for automated Q&A generation
By
–
One handy use case for this tool is to auto generation of a Q&A section based on a given context. Here it becomes more useful considering a Gemini 1.5 Pro's huge context window.
-
Technical limitations of NotebookLM document context window
By
–
For example, you can give TestingCatalog URL and explore the content via NotebookLM. It feels like it is trying to scrape the date up until it reaches a certain context window limit. So you never know how complete and reliable it would be
-
Google NotebookLM adds URL support for source material
By
–
Google's NotebookLM now can take any URL as a source. It would let you chat and take notes based on the conversation with the date from this source.