kalqlate

Perpetually kalqlating and preparing for future life increasingly beyond imagination. #Technology #AI #Robotics #SystemsArchitecture #SoundDesign #Music #AIArt

[kalqlate @kalqlate1]

@MusaM5834 REJECTED! Left feeling dejected.

[kalqlate @kalqlate1]

@MusaM5834 Perhaps even deeper than that. With other monkeys watching... oh, the shame and embarrassment that you gave your intimacy, trust, care, and concern and got none in return.

[Andrej Karpathy @karpathy]

You can tell the RL is done properly when the models cease to speak English in their chain of thought

[Chubby♨️ @kimmonismus]

Soon AI will be used to adapt graphics in games in real time. The possibilities are unimaginable because everything will be possible.

[Interesting AF @interesting_aIl]

Imagine being stoned in a place like this

[AK @_akhaliq]

Tencent presents GameGen-O

Open-world Video Game Generation

We introduce GameGen-O, the first diffusion transformer model tailored for the generation of open-world video games. This model facilitates high-quality, open-domain generation by simulating a wide array of game engine features, such as innovative characters, dynamic environments, complex actions, and diverse events. Additionally, it provides interactive controllability, thus allowing for the gameplay simulation. The development of GameGen-O involves a comprehensive data collection and processing effort from scratch. We collect and build the first Open-World Video Game Dataset (OGameData), amassed extensive data from over a hundred of next-generation open-world games, employing a proprietary data pipeline for efficient sorting, scoring, filtering, and decoupled captioning. This robust and extensive OGameData forms the foundation of our model's training process. GameGen-O undergoes a two-stage training process, consisting of foundation model pretraining and instruction tuning. In the first phase, the model is pre-trained on the OGameData via the text-to-video and video continuation, endowing GameGen-O with the capability for open-domain video game generation. In the second phase, the pre-trained model is frozen, and we fine-tuned using a trainable InstructNet, which enables the production of subsequent frames based on multimodal structural instructions. This whole training process imparts the model with the ability to generate and interactively control content. In summary, GameGen-O represents a notable initial step forward in the realm of open-world video game generation via generative models. It underscores the potential of generative models to serve as an alternative to rendering techniques, which can efficiently combine creative generation with interactive capabilities.

[kalqlate @kalqlate1]

@AngryTomtweets LOL! I think I watched through about four times before realizing it was repeating. No matter... I watched another four times after. 😄

[Angry Tom @AngryTomtweets]

8 AI videos that will leave you speechless.

1. "Return"

[kalqlate @kalqlate1]

🤣🤣🤣

[Ai🩵Ai @L0V3Ai_Ultimate]

@kalqlate1 Oh Yeah! 🤣😂🤣 We Got this!

[kalqlate @kalqlate1]

The stuff dreams AND nightmares are made of. 🤣

[Rishabh @Rixhabh__]

3.

[Everett World @WorldEverett]

🧵 Luma Labs Dream Machine 1.5 (thread)

Dream Machine 1.5 is insanely good.

I’ve got 8 amazing new videos created with prompts👇

1/8 Prompt: small city in northwest US, car driving in the snow, reflections in the windshield, hyperspeed

[kalqlate @kalqlate1]

@WorldEverett Hee hee! I think these image/video models sometimes get confused about what side to place oncoming traffic due to the training data containing videos from all over the world, which includes different driving side rules in different countries. The same thing WAS true for fingers.

[Luma @LumaLabsAI]

Dream Machine 1.5 is here 🎉 Now with higher-quality text-to-video, smarter understanding of your prompts, custom text rendering, and improved image-to-video! Level up. https://t.co/G3HUEBE2ng
#LumaDreamMachine

[Everett World @WorldEverett]

Midjourney, Gen-3, and Udio are mind-blowing together.

Last year was all about images, this year is about videos, and it's such a joy to share this with you!

Images: Midjourney v 6.1
Video: Gen-3 by Runway
Music: Udio 1.5
Editing: CapCut

[Everett World @WorldEverett]

@agoraitconsulti @Kyrannio I loved it too! To the point that I generated the entire song, and I've been listening to it whole day, haha :D Enjoy!

[kalqlate @kalqlate1]

@WorldEverett @agoraitconsulti @Kyrannio Truly, one of the best! In my 2nd year of learning music prod. Started back at first use of ChatGPT - I saw the inevitable end of my dev/architecture career, knew AI music would take over, but decided to dive in and have fun anyway. 😂 Having great fun, but Udio is just so good!

[kalqlate @kalqlate1]

@agoraitconsulti @WorldEverett @Kyrannio "These days I also only listen to the music I generated as well :)" -- For several months now on YouTube, I've gotten pushback in the producer community by stating this would become the default case for millions, leaving increasingly less ear-share for human-produced music.

[kalqlate @kalqlate1]

@mara_hopewell Thanks! This post was just to mark the occasion for future reference. I'll post more in the near future and along the way of our continual progress.

[kalqlate @kalqlate1]

🥳Funded and in development - smart, organic, any-domain machine - highly-abstracted, intuitive, flexible, workflow/info mgmt, targeting traditional and metaverse. Designed over 1.5 years, now with deep AI integration for custom/known domain knowledge and reasoning. EXCITING!🥳