cturan/llama.cpp

mirror da https://github.com/cturan/llama.cpp

Autore	SHA1 Messaggio	Data
compilade	f98eb31c51 convert-hf : save memory with lazy evaluation (#7075)	1 anno fa
Brian	a2ac89d6ef convert.py : add python logging instead of print() (#6511)	1 anno fa
Kerfuffle	91f6499393 Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)	2 anni fa
Kerfuffle	34b0a08207 gguf-py: Refactor and allow reading/modifying existing GGUF files (#3981)	2 anni fa