๐ retweeted
Replying to @elonmusk @Scobleizer
Itโs crazy to think we are engaging with a dude who will be in history books 500 years from nowโฆWhile we are all forgotten about within 50 lol
Wow! Look at whay happens when you start messing with trillion $$ corporations profits. They actually listen. Funny how that works.
eurogamer.net/skate-mods-livโฆ
๐ retweeted
"I don't have a GPU" is officially dead ๐คฏ
You can now run 70B model on a single 4GB GPU and it even scales up to the colossal Llama 3.1 405B on just 8GB of VRAM.
AirLLM uses "Layer-wise Inference." Instead of loading the whole model, it loads, computes, and flushes one layer at a time
โ No quantization needed by default
โ Supports Llama, Qwen, and Mistral
โ Works on Linux, Windows, and macOS
100% Open Source.
Readers added context they thought people might want to know
AirLLM enables the claimed low VRAM usage via per-layer disk streaming but requires reloading the full model from storage for each token, yielding speeds like ~20s/token for 8B models and minutes per token for 70B+ per tests and project reports.
kunalvohra.com/blog/airllm-onโฆ
github.com/lyogavin/airllโฆ
NVIDIA Shield TV Pro is now $299, nearly seven years after launching at $199 - What a fucking joke. Nearly 10 year old technology being raised in price. Get absolute fucked lmfaooo it just doesnt end... my question is when is the populous going to break.. share.google/rw6OQlU8mPrC2tjโฆ
๐ retweeted
Replying to @ChrisRushton15 @TheWyteRabbit1
For the first time, Japan has more than 100,000 centenarians | CNN
Meanwhile in 'Merica ๐ซข
cnn.com/2026/09/15/asia/japaโฆ
๐ retweeted
Indian factory workers wearing head-mounted cameras to record hand movements for training AI systems
๐ retweeted
This is Nigeria ๐ณ๐ฌ.
Full documentary on modern pirates attacking powerful speedboats, abducting foreign nationals and demanding huge ransoms for their release.



















