25 Commits

Author	SHA1	Message	Date
csunny	d67a6a642a	fix: num_gpus referenced error for mps + cpu	2023-08-03 16:52:39 +08:00
FangYin Cheng	a4574aa614	feat: Support vicuna-v1.5 and WizardLM-v1.2	2023-08-03 14:14:29 +08:00
FangYin Cheng	d8a4b776d5	feat: Support 8-bit quantization and 4-bit quantization for multi-gpu inference	2023-08-02 19:29:59 +08:00
zhanghy-sketchzh	00d24101f3	support multi gpus	2023-06-14 00:22:02 +08:00
csunny	fe8291b198	feature: guanaco stream output	2023-06-04 20:38:34 +08:00
csunny	09308bcdf0	fix: guanaco model	2023-05-31 14:13:12 +08:00
csunny	16c6986666	fix: lint	2023-05-30 19:11:34 +08:00
csunny	ea334b172e	feature: add model server proxy	2023-05-30 17:16:29 +08:00
yihong0618	b098a48898	ci: make ci happy lint the code, delete unused imports ... Signed-off-by: yihong0618 <zouzou0208@gmail.com>	2023-05-24 18:43:04 +08:00
yihong0618	60ecde5892	fix: can not answer on mac m1-> mps device	2023-05-24 12:33:41 +08:00
csunny	f52c7523b5	llms: fix	2023-05-21 14:54:16 +08:00
csunny	ce72820085	llms: add mps support	2023-05-21 14:48:54 +08:00
csunny	4302ae9087	Add: multi model support	2023-05-18 15:44:29 +08:00
csunny	6d76825a10	rm fschat relay	2023-05-11 10:59:08 +08:00
csunny	fd8bc8d169	modelLoader use singleton	2023-05-10 10:53:48 +08:00
csunny	bfbbf0ba88	update conversation	2023-05-09 21:48:47 +08:00
csunny	d746086694	adjust project content	2023-05-08 00:34:36 +08:00
csunny	539e98f1dc	fork file replace import	2023-05-07 05:14:43 +08:00
csunny	eca14bc038	fix load model gpu oom	2023-04-29 23:02:13 +08:00
csunny	acf9dbbd82	fix problem	2023-04-29 21:50:47 +08:00
csunny	0767537606	add vicuna embedding	2023-04-29 18:28:42 +08:00
csunny	e5ffb6582c	a demo	2023-04-28 23:53:29 +08:00
csunny	0861a09a00	init model and tokenizer	2023-04-28 22:18:08 +08:00
csunny	c72ae1a87f	model: add model file	2023-04-28 22:04:37 +08:00
csunny	38f57e157c	init	2023-04-28 21:59:18 +08:00