09-09-2026, 03:54 PM
Interesting, I hadn't tried that. I ended up in about the same place using PyTorch. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
|
throughput per watt reality check with llama.cpp
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
throughput per watt reality check with llama.cpp - by griffinloom66 - 09-09-2026, 01:07 AM
Re: throughput per watt reality check with llama.cpp - by batchwire - 09-09-2026, 05:09 AM
Re: throughput per watt reality check with llama.cpp - by homelabhenry - 09-09-2026, 04:07 PM
Re: throughput per watt reality check with llama.cpp - by macronest - 09-09-2026, 07:16 AM
Re: throughput per watt reality check with llama.cpp - by turboscope - 09-09-2026, 03:54 PM
Re: throughput per watt reality check with llama.cpp - by thetafield - 09-10-2026, 04:38 PM
Re: throughput per watt reality check with llama.cpp - by forgemixer - 09-11-2026, 01:21 AM
Re: throughput per watt reality check with llama.cpp - by emberpeak - 09-11-2026, 04:05 AM
Re: throughput per watt reality check with llama.cpp - by bytegarden73 - 09-11-2026, 05:07 AM
|
Users browsing this thread: 1 Guest(s)