Damus
Open in Damus
Dave · 7w
We’re in the mainframe/timeshare era of LLM’s right now. The PC era is coming. There were demand induced shortages back then too: https://www.nytimes.com/1988/03/12/us/shortage-of-memory-chips-ha...
 Chris L-W👻🚫 profile picture
Chris L-W👻🚫 @Chris L-W👻🚫
@nprofile1q... I think RAM, SSD, and other prices will be coming down after the IPOs. Investors might deal with low/no profitability, but shareholders will not.

So, the big AI companies will need to scale back on datacenter plans and such.
1
Dave · 7w
nostr:nprofile1qy2hwumn8ghj7un9d3shjtnyd968gmewwp6kyqpqk3yvm2djkxkh46jt0nv72hkz5vqam059dy7xv266lu9x5dhfspes6ek5s4 My recent experience with Qwen3.6 35B running locally on a SGX Spark tells me that the local models are virtually on par with the hyperscale models for many tasks like coding. A develope...