Replies: 2 comments
-
Seems really interesting. Microsoft developed it as a way to save $ when interacting with paid servers like OpenAI, but I'm wondering if it could be beneficial for llama.cpp. MS claim the method can achieve prompt compressions of up to 20x and get ~same response from the LLM. I'd certainly love to have 160K effective context for a mistral model! |
Beta Was this translation helpful? Give feedback.
0 replies
-
Can you imagine? Current 32K context models upgraded to 640K ..wow. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
https://github.com/microsoft/LLMLingua/tree/main
It tried to compress prompt and document so can be much smaller, I just wondered if we can use it with llama cpp (gguf)
Beta Was this translation helpful? Give feedback.
All reactions