Difference between speakleash and DavidAU's Bielik GGUF
#1
by
mrsh7
- opened
How is this GGUF different from bielik-7b-instruct-v0.1.Q6_K.gguf from the speakleash/Bielik-7B-Instruct-v0.1-GGUF repository? I've done some testing and I get essentially the same results on them.
We added this one to our repository only because of the "instruct" component. We also added it for back up purposes for our use cases. At is uses the standard LLAMA.CPP (and default quant settings) it is exactly the same or should be... unless changes occurred in llama.cpp between the time it was originally quanted and when we did it. As LLAMA.CPP is under daily development some changes do occur in both quanting and actual use of the model via llama.cpp and it's downstream apps like "WebUI",. LmStudio, Ollama and so on.
mrsh7
changed discussion status to
closed
Thank you