Realtime 3D rendering engine driven by a diffusion model
category: code [glöplog]
rloaderro: if you want image generator as versatile as Midjourney, they you need billions of params..but maybe you can narrow down the focus, say to terrain features. One direction I am willing to explore is model compression techniques, like pruning and quantization (and many others). This not only makes the model smaller, but also faster for inference and very much touches demoscene size-coding spirit. 64kb might still tight though ;P
Sounds cool! Hope you can put it into practice on the big screen soon :)
Yeah, right, for zealots waving "fuckai" banners to talk shit about it - that definitely gives plenty of motivation to do it.
"zealots" vs "NPCs"
The problem is zealots only see two sides of the coin.
To repeat myself (tenth time), I'm not at the fore-front of AI hype and I dislike the trend of putting "AI" everywhere you can imagine, but I do see a value in reasonable use of modern ML techniques.
To repeat myself (tenth time), I'm not at the fore-front of AI hype and I dislike the trend of putting "AI" everywhere you can imagine, but I do see a value in reasonable use of modern ML techniques.
You two get a fucking room.
Another team published a counter strike model that actually runs in realtime on a normal PC (3090 or 4090 required, also I havent tried it yet).
https://github.com/eloialonso/diamond/tree/csgo
It seems they have also trained the model on a singl RTX4090!
https://github.com/eloialonso/diamond/tree/csgo
It seems they have also trained the model on a singl RTX4090!