OpenAI's New GPT Model Release (0:43) Discussion on the new GPT model (01 pro) and its improvements. Comparison of GPT Models (0:53) Detailed comparison of the new model with previous versions using various benchmarks. Pricing and Access to GPT Models (2:50) Information on the cost and access options for different GPT models. Google's Game-From-Screenshot AI (Cheny 2) (3:30) Brief mention of an AI that can create a video game from a single screenshot. Flow State AI Tool (3:50) Introduction to Flow State, an AI tool for brainstorming and generating images. Hunan Video Model (5:17) Discussion on the Hunan video model and its capabilities, including how to run it locally. Cataloging Home Items with Gemini (7:41) A project using Gemini to catalog and identify objects in a home. Image-to-Video Control Research by Bite (8:12) Research on controlling video elements like camera movement and object motion. Runway's Key Framing Prototype (9:36) Introduction to a new key framing prototype by Runway for video creation.
In finde Du brauchst Dich für solche Dinge wie das Mikro nicht entschuldigen. Dein Content ist gut, wie er ist und spricht für sich. Deine Show, im wahrsten Sinne des Wortes. Du kommst selbst in einer fremden Sprache liebenswürdig rüber und wirst gemocht, wie man ja an den Abonnenten sieht. Die Arbeit, die Du machst bringt den Menschen Wert. Und Du machst sie gut. Kein Grund für Entschuldigungen, also. Einfach Deinem Bauchgefühl folgen, es hat Dich immerhin schon bis hier gebracht. Keep it up, Olivio. Und damit meinte ich nicht das Mikro in der Hand. :)
Danke, das ist echt nett von dir. ich wollte mich eigentlich nicht entschuldigen, sondern dachgte das die Story irgendwie witzig ist und ich wollte mal was persönliches mit rein bringen :)
@@OlivioSarikas Ja, man kann immer mal was Neues ausprobieren. Aber um "persönlich" brauchst Du Dir keine Sorgen machen. Deine Persönlichkeit ist es, was den Kanal ausmacht. Sowieso schon. 👍
Great video! The new video model especially looks interesting and competitive with online platforms finally. Though I suppose LTX is also not bad now either.
Yeah.. shocked at either how much paywalled it all became, or how much censored/lobotimized (even more) it would be, or both. I'm rather pessimistic about it tbf.
What was predicted by a few, is slowly becoming a reality. AI models will become less and less cheap, and more and more expensive, if you want to get the decent results. Time to pay back the investors. It was not an endless party
sadly yes. but to be honest i get everything i need from the free version. and consumers don't have endless money. if this was a fully integrated assisteant that can do more than being a multi model llm in a box, i might consider it.
I'm with Olivio, for images I pretty much get what I want already and while I've dabbled with video, it's not really what I'm interested in. Text and ChatGPT stuff has no interest to me whatsoever. The point about not having endless money is also very valid. If they aren't careful they could end up with a 'live service' situation in gaming, where they all want your time and money but there's too many services and nobody wins.
Will see. But otherwise, I think they will be focusing more and more on the features that actually make the best results, like video to video, expand mode in runway etc. Because if they want people to pay for the service, they need to deliver actual value, and less beta testing features. And also making more and more separation between features for consumer versus business plans.
@@vladartiomav2473 Yes, I can agree with that. It seems like beta testing now because so much is developing fast with people jumping on the bandwagon, especially with video. Image generation has already been through that and now it's more settled, at some point the same will apply to video at some point.
I can't wait for all these services to paywall themselves into oblivion. At some point we will see the real cost of all these services and what the value of real animator and artists is.
OpenAI's New GPT Model Release (0:43)
Discussion on the new GPT model (01 pro) and its improvements.
Comparison of GPT Models (0:53)
Detailed comparison of the new model with previous versions using various benchmarks.
Pricing and Access to GPT Models (2:50)
Information on the cost and access options for different GPT models.
Google's Game-From-Screenshot AI (Cheny 2) (3:30)
Brief mention of an AI that can create a video game from a single screenshot.
Flow State AI Tool (3:50)
Introduction to Flow State, an AI tool for brainstorming and generating images.
Hunan Video Model (5:17)
Discussion on the Hunan video model and its capabilities, including how to run it locally.
Cataloging Home Items with Gemini (7:41)
A project using Gemini to catalog and identify objects in a home.
Image-to-Video Control Research by Bite (8:12)
Research on controlling video elements like camera movement and object motion.
Runway's Key Framing Prototype (9:36)
Introduction to a new key framing prototype by Runway for video creation.
Thank you, i made it a sticky and put it into the video info
Genie 2 looks cool.
I love LTX v with Add LTX Latent Guide to enable image to video and STSG to enhance the video output.
In finde Du brauchst Dich für solche Dinge wie das Mikro nicht entschuldigen. Dein Content ist gut, wie er ist und spricht für sich. Deine Show, im wahrsten Sinne des Wortes.
Du kommst selbst in einer fremden Sprache liebenswürdig rüber und wirst gemocht, wie man ja an den Abonnenten sieht. Die Arbeit, die Du machst bringt den Menschen Wert.
Und Du machst sie gut. Kein Grund für Entschuldigungen, also. Einfach Deinem Bauchgefühl folgen, es hat Dich immerhin schon bis hier gebracht. Keep it up, Olivio.
Und damit meinte ich nicht das Mikro in der Hand. :)
Danke, das ist echt nett von dir. ich wollte mich eigentlich nicht entschuldigen, sondern dachgte das die Story irgendwie witzig ist und ich wollte mal was persönliches mit rein bringen :)
@@OlivioSarikas Ja, man kann immer mal was Neues ausprobieren. Aber um "persönlich" brauchst Du Dir keine Sorgen machen. Deine Persönlichkeit ist es, was den Kanal ausmacht. Sowieso schon. 👍
@@sdhpCH danke :)
@@sdhpCHOlivio, ich sag einfach mal "what he said" plus geiles Shirt. 😂
Outstanding shirts👍
holding the mic gives a unique look
Great video! The new video model especially looks interesting and competitive with online platforms finally. Though I suppose LTX is also not bad now either.
In my testing, I found HuYuan video to far superior to LTX if you use the Sage_attention flag.
Low-Tech but high quality
Part of me really wants to step away from AI for like 5-6 months so I can come back truly shocked
Other part of you wants to stay because the next big could be 5-6 days or even hours ahead?! 😊
Yeah.. shocked at either how much paywalled it all became, or how much censored/lobotimized (even more) it would be, or both. I'm rather pessimistic about it tbf.
What was predicted by a few, is slowly becoming a reality. AI models will become less and less cheap, and more and more expensive, if you want to get the decent results.
Time to pay back the investors. It was not an endless party
sadly yes. but to be honest i get everything i need from the free version. and consumers don't have endless money. if this was a fully integrated assisteant that can do more than being a multi model llm in a box, i might consider it.
I'm with Olivio, for images I pretty much get what I want already and while I've dabbled with video, it's not really what I'm interested in. Text and ChatGPT stuff has no interest to me whatsoever.
The point about not having endless money is also very valid. If they aren't careful they could end up with a 'live service' situation in gaming, where they all want your time and money but there's too many services and nobody wins.
Will see. But otherwise, I think they will be focusing more and more on the features that actually make the best results, like video to video, expand mode in runway etc.
Because if they want people to pay for the service, they need to deliver actual value, and less beta testing features.
And also making more and more separation between features for consumer versus business plans.
@@vladartiomav2473 Yes, I can agree with that. It seems like beta testing now because so much is developing fast with people jumping on the bandwagon, especially with video. Image generation has already been through that and now it's more settled, at some point the same will apply to video at some point.
I can't wait for all these services to paywall themselves into oblivion. At some point we will see the real cost of all these services and what the value of real animator and artists is.
I'm having some real difficulty finding local ai apps in the windows app store.
2025 is the most important year in human history.
We are at the crossroads.
I see ppl talking about changes in Leonardo but how come everybody is absolutely sleeping on Invoke?
invoke was too slow with their updates and because of that didn't stay relevant. But i will do a video about it some time soon
The only wild thing about AI in 2025 will be how 32GB of vram is not enough while it gets you the same quality of SDXL but 100x more taxing.
9:18 it's a hedgehog.
Definitely came from canada, cuz hi dont think he knew about salinas