This is epic
🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to…
I think i took a big risk and probably made a very bad decision yesterday due to being too optimistic. let's see.
tanhai mai aur tum andheke rasto me sharmana ghabrana kaisa hai apno me meri jaan mera mann. tera mann mera mann tera mann yun hi milta hai ab dil se dil milao aur thoda sa muskurao zara chehra to dikhao aur thoda sa muskurao aur thoda sa muskurao.
duniya ke gam bhulao aur thoda sa muskurao khushboo ka foolo se sadiyo ka rishta hai geeto ka sur jaise rang dhal jaata hai tera mann tera mann mera mann. tera mann. vaise hi milta hai ab dil se dil milao aur thoda sa muskurao zara chehra to dikhao aur thoda sa muskurao...
im always lagging behind one more moment to catch a moment actually id like to walk a little more slowly the light looks nice on the water today let's just stand here for a while and stare oh gosh im probably late again
that's why yann le cun is the goat, he published his opinion paper which had the architectures using JEPA/world models as RL environments to do step by step planning, tasks etc.
the highest fidelity simulators of our world are almost here you’re preparing to RL against them, right…? surely you realized the point was to use the world models as RL envs… didn’t you, anon? you’re preparing… aren’t you?
LLMs or ai can't be an all purpose things by themselves, by being just neurons. atleast nowhere in near future. it has to be neurosymbolic, an agent - ai should just be a planner - planning and ordering tools to do things overseeing everything.
I think gpt 5 is a big breakthrough, in a real sense. always knew this scaling law can't be squeezed forever, but the race of llms before this was just improving a bit on benchmarks over others. now gpt5 with tool calling is to shift focus to making llms actually useful.
honestly i never used to like chatgpt that much, for sometime now.. even for coding tasks , it gives best error fixing, and context lengths are crazy on chat and seems like almost unlimited chat use no restriction
We asked a question to our self.... Why does reasoning model are only supposed to reason on top, why not reason mid response. Kinda like how claude do it using think tool. Introducing Dhanistha-2.0-preview World's First Intermediate reasoning model
United States 趨勢
- 1. #VSFashionShow 231K posts
- 2. #youtubedown 5,994 posts
- 3. quen 14.4K posts
- 4. Missy 7,589 posts
- 5. madison 45.8K posts
- 6. jihyo 95.4K posts
- 7. #VictoriasSecretFashionShow 3,085 posts
- 8. Karol G 33.9K posts
- 9. Angel Reese 18.2K posts
- 10. #KaneAI N/A
- 11. tzuyu 88.9K posts
- 12. bella hadid 23.1K posts
- 13. #ARWINGS 5,345 posts
- 14. Nancy 112K posts
- 15. Birdman 2,550 posts
- 16. Anok 11.1K posts
- 17. Candice 14.4K posts
- 18. Argentina 437K posts
- 19. nayeon 77.1K posts
- 20. Keegan Murray 1,171 posts
Something went wrong.
Something went wrong.