Yes. This heavily subsidized LLM inference usage will not last forever. We have already seen cost cutting for some models. A model starts strong, but over time the parent company switches to heavily quantized versions…
GLM-4.5-Air AWQ Q4 is fantastic all around (including coding), and can run on 4 RTX 3090s easily.
I definitely do not make a living from it, but after two years, my Bitcoin on-chain data and metrics API service has started to generate some revenue. It's basically a service where you can download highly processed…
NOSTR!
Blue light blocker glasses to wear at nigth. Changed my life in so many ways. I thought I slept well before, but boy I was so wrong!
What does the "Satoshi Nakamoto 1A1zP1eP5QGefi2DMPTfTL5SLmv7DivfNa" on page 1 means?
Yes. This heavily subsidized LLM inference usage will not last forever. We have already seen cost cutting for some models. A model starts strong, but over time the parent company switches to heavily quantized versions…
GLM-4.5-Air AWQ Q4 is fantastic all around (including coding), and can run on 4 RTX 3090s easily.
I definitely do not make a living from it, but after two years, my Bitcoin on-chain data and metrics API service has started to generate some revenue. It's basically a service where you can download highly processed…
NOSTR!
Blue light blocker glasses to wear at nigth. Changed my life in so many ways. I thought I slept well before, but boy I was so wrong!
What does the "Satoshi Nakamoto 1A1zP1eP5QGefi2DMPTfTL5SLmv7DivfNa" on page 1 means?