Check out Vozo AI at: Use code【aiadvantage】to get extra 30 points, only for the first 100 users. It’s been a big week for those of us excited about the future of AI, as we’ve received huge updates to GPT-4o, a glimpse into the future of AI agents from Mistral, a free and open source AI image generator that creates perfect text in images, and way…
Igor, greetings from SF! Thanks for sharing the AI tool updates. You've given me several to check out! 👍
For Gemini, how did you get your interface to look like that, and be able to select the different models?
Exciting updates! With SmythOS, you can fully profit from these AI moves and incorporate them seamlessly into your projects. #AIUpdates #SmythOS
Hey Igor, Gemini 1.5 Pro Experimental is amazing. 2 million context window and very sharp and free, no timeouts. It made my day when I noticed it had like a Windows UAC feature to turn nagging way down and policy guard as well.
Why is dall-e not part of this? Amazing cool consice review. Good job!
Thanks buddy
Can my ai channel get shout out
Consistent json output is massive for devs!!!
I tried putting input and output pairs into Claude projects knowledge base, but it had unintended consequence of poisoning new prompts with some of the input/output training data
Best thumbnails on YouTube. 👏🔥 And I LOVE how you include ALL of the links to the platforms that you talk about. Much appreciated! (The Olympics product placement 😆 a bit cringe dude sorry ) Totally agree re Flux. It’s not mid journey. Glif is amazing for prompt enhancement though. Thank you!
How do you these info? You bundle it for us but you obviously had to gather this individually. How?
Yoh I'm a flux fan and you undervalued my buddy 😢❤
yea the problem with the video generation is that its going to do its own thing and that could be random, so if you have a specific output in mind then its not very useful.. But if a community just built a library of stock footage then we could find the AI video that we want if there's a big enough library.
Thanks for the video! By the way, 'Le chat' in French means 'The cat,' and it's pronounced 'Luh sha,' without pronouncing the 't' at the end 😹. Quick update: Flux has released a new model called 'FLUX.1-merged' 😉
are you using flux dev though?
On your Gen-3 experiments, the reason you were not getting what you expected was likely that you were feeding it the final frame, then giving it instructions on what to do after the last frame. The yellow paint wouldn't replace the clean water at the end, it would have to start as yellow paint and become the clean water.
Analogous, Incorrect Example: [final frame is an image of an intact window], User: "Break the window." [Output: but… but, the window is intact in the last frame, how can I break the window into a intact state? *explodes*]
That was massive, heaps of news
Great update on the latest AI developments! The comparison between Flux and MJ was insightful, and the discussion on AGI was thought-provoking. Keep up the excellent work in breaking down complex AI concepts for us. ZYTHI STUDIO https://www.youtube.com/@ZYTHI ,Explore AI-powered video creation and more on my channel!
seriously? at 13:25 is there still who gives a shi*ft & watch the dirty Olympics? u disappointed me 🤧
good video btw, thank u.
the Game of Thrones AGI video was awesome 😂
Great video, thanks for the brilliant research
That's not a beaver! 🦫🦫😳
Flux looks better imo… 100% , and is version. 1 xd
Which version of Flux did you use for the comparisons?
never stop doing this week's news content ❤