3 ms·
I wonder how many years it'll take for the API token cost to exceed the money spent on ram.
by rjh29 5mo ago
I wonder how many years it'll take for the API token cost to exceed the money spent on ram.
- zozbot234 5mo agoThe DS4 folks are unofficially testing ways to run the model with lower performance on lower-RAM machines. Similar efforts are going on with llama.cpp. The results are a bit of a challenge, prefill time tends to explode which is a limitation if you care about agentic workflows.