OpenSource/ollama

Auteur	SHA1 Message	Date
Michael Yang	b599946b74 add format bytes	il y a 1 an
Bruce MacDonald	d06bc0cb6e enable q8, q5, 5_1, and f32 for linux gpu (#699)	il y a 1 an
Bruce MacDonald	86279f4ae3 unbound max num gpu layers (#591)	il y a 1 an
Bruce MacDonald	4cba75efc5 remove tmp directories created by previous servers (#559)	il y a 1 an
Michael Yang	7dee25a07f fix falcon decode	il y a 1 an
Bruce MacDonald	09dd2aeff9 GGUF support (#441)	il y a 1 an
Bruce MacDonald	42998d797d subprocess llama.cpp server (#401)	il y a 1 an
Michael Yang	b25dd1795d allow F16 to use metal	il y a 1 an
Michael Yang	304f2b6c96 add 34b to mem check	il y a 1 an
Michael Yang	a894cc792d model and file type as strings	il y a 1 an
Michael Yang	e26085b921 close open files	il y a 1 an
Michael Yang	6de5d032e1 implement loading ggml lora adapters through the modelfile	il y a 1 an
Michael Yang	d791df75dd check memory requirements before loading	il y a 1 an
Michael Yang	020a3b3530 disable gpu for q5_0, q5_1, q8_0 quants	il y a 1 an
Michael Yang	fccf8d179f partial decode ggml bin for more info	il y a 1 an