10-12-2026, 06:47 AM
I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model.
retrieval before vibes
|
prefix caching in llama.cpp for Kimi
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
prefix caching in llama.cpp for Kimi - by macroforge - 10-08-2026, 03:58 AM
Re: prefix caching in llama.cpp for Kimi - by vrampilot - 10-08-2026, 04:21 PM
Re: prefix caching in llama.cpp for Kimi - by mangotinker73 - 10-08-2026, 05:15 PM
Re: prefix caching in llama.cpp for Kimi - by visionyard23 - 10-09-2026, 01:27 AM
Re: prefix caching in llama.cpp for Kimi - by embershed99 - 10-09-2026, 06:10 AM
Re: prefix caching in llama.cpp for Kimi - by latentcraft - 10-09-2026, 06:26 AM
Re: prefix caching in llama.cpp for Kimi - by macropilot - 10-12-2026, 06:47 AM
|
Users browsing this thread: 1 Guest(s)