Trendaavat aiheet
#
Bonk Eco continues to show strength amid $USELESS rally
#
Pump.fun to raise $1B token sale, traders speculating on airdrop
#
Boop.Fun leading the way with a new launchpad on Solana.
Awni Hannun
AI @apple
Cool exploration of distributed inference on heterogenous hardware. Including Apple silicon with MLX:
2.8. klo 21.00
Intelligence has been locked in walled gardens. Today, we’re opening the gates.
Parallax now runs in Hybrid mode, with Macs and GPUs serving large models together in a truly distributed framework.
15,4K
Super intelligence in your pocket will happen. And if you want to accelerate the timeline hit me up.
We’re working on this at every layer with MLX.
6.8. klo 07.48
someday soon something smarter than the smartest person you know will be running on a device in your pocket, helping you with whatever you want.
this is a very remarkable thing.
61,28K
Super intelligence in your pocket is going to happen. And if you want to accelerate the timeline hit me up.
We’re working on this at every layer with MLX.
6.8. klo 07.48
someday soon something smarter than the smartest person you know will be running on a device in your pocket, helping you with whatever you want.
this is a very remarkable thing.
206
I only just started playing with it, but the model seems great so far.
But it also has some implementation idiosyncrasies:
- A new chat protocol
- Only available in fp4 quantization
- Attention sink which kind of breaks fused attention
Open models move fast and I wonder how much time to invest in supporting these features? Will OpenAI open source more models?
6.8. klo 12.43
OpenAI's new 120B MoE runs nicely in mlx-lm on an M3 Ultra.
Running the 8-bit quant:
178
Johtavat
Rankkaus
Suosikit