github/DB-GPT
1
0
Fork 0
mirror of https://github.com/csunny/DB-GPT.git synced 2026-07-27 08:48:41 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,105 Commits 66 Branches 54 Tags
2ea0efc19dbb6ac571e371cf12d65ca320079562
Commit Graph

182 Commits

Author SHA1 Message Date
FangYin Cheng
5dfe611478 feat(ChatKnowledge): Add custom text separators and refactor log configuration 2023-09-28 11:54:58 +08:00
yihong0618
76fdf52b0a fix: another way 2023-09-23 14:46:26 +08:00
yihong0618
24e6d16a09 fix: proxyllm do not need controller_addr 2023-09-22 16:47:34 +08:00
FangYin Cheng
28c24516a2 chore: Add _clear_model_cache function to clear model cache 2023-09-21 18:19:19 +08:00
FangYin Cheng
896af4e16f chore: fix shutdown error when not install torch 2023-09-21 12:20:24 +08:00
FangYin Cheng
33506b062b feat(core): Multi-module dependency splitting 2023-09-20 17:31:56 +08:00
FangYin Cheng
2d4d513eb5 chore: rename component 2023-09-19 10:23:39 +08:00
FangYin Cheng
ae34be23fd chore(model): new required field for supported models 2023-09-19 09:58:48 +08:00
FangYin Cheng
78553477a9 feat(model): new worker manager factory componet 2023-09-19 00:30:15 +08:00
FangYin Cheng
5aa9cb455e feat(model): support openai embedding model 2023-09-15 18:08:55 +08:00
FangYin Cheng
8da81486f5 fix(model): Fix remote embedding model error in some case 2023-09-14 20:11:41 +08:00
FangYin Cheng
f304f9709d fix(model): Fix the bug that the webserver cannot return model instances 2023-09-14 12:36:18 +08:00
FangYin Cheng
7b64c03d58 feat(model): support InternLM 2023-09-14 11:32:34 +08:00
FangYin Cheng
581cf361bf feat(model): multi-model supports embedding model and simple component design implementation 2023-09-13 12:14:21 +08:00
FangYin Cheng
27d7f9f6b9 chore: fix pylint 2023-09-11 17:28:22 +08:00
FangYin Cheng
986ada3aeb feat(model): supports the deployment of multiple models through the API and add the corresponding command line interface 2023-09-11 17:15:20 +08:00
FangYin Cheng
4744d2161d fix(model): fix proxy llm with fix model from CONFIG 2023-09-08 19:46:50 +08:00
FangYin Cheng
a8846c40aa feat(model): Unified Deployment Mode with multi-model and add command line chat with model 2023-09-08 17:52:48 +08:00
FangYin Cheng
b6a4fd8a62 feat: Multi-model support with proxyllm and add more command-cli 2023-09-05 11:26:24 +08:00
aries_ckt
858a1caefd fix:worker_register_host parameter does not work 
Close #526
 2023-09-03 18:14:51 +08:00
FangYin Cheng
e5bbd0bd86 feat: Command-line tool with knowledge repository initialization 2023-09-01 18:21:22 +08:00
FangYin Cheng
f19551a7cd feat: Optimize code import time 2023-09-01 10:40:18 +08:00
FangYin Cheng
e4dd6060da feat: Command-line tool design and multi-model integration 2023-08-31 17:22:31 +08:00
FangYin Cheng
dd86fb86b1 feat: Multi-model command line 2023-08-30 11:07:35 +08:00
FangYin Cheng
d467092766 merge multi-model and ChatExcel 2023-08-29 22:54:18 +08:00
csunny
e058511b4e pylint: lint code format 2023-08-23 17:02:54 +08:00
csunny
5f53463221 fix: gpt4all output bug 2023-08-23 16:47:59 +08:00
FangYin Cheng
efd17593ce fix: Fix alpaca prompt error 2023-08-18 17:18:01 +08:00
FangYin Cheng
909bb263a4 feat: Add examples and modify deploy documents 2023-08-18 15:12:37 +08:00
FangYin Cheng
b5fd5d2a3a feat: Support llama.cpp 2023-08-15 19:00:08 +08:00
csunny
743863d52b fix: set num_gpus reference for mps + cpu 2023-08-03 16:54:58 +08:00
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
bceb609cf6 feat: Modify config for quantization and doc 2023-08-02 19:29:59 +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
xuyuan23
dea677004a fix: simply the code and optimization 
simply the code and optimization
 2023-07-31 19:19:39 +08:00
xuyuan23
1ff4cfdc42 feat: support bard proxy server 
support bard proxy server

Close #384
 2023-07-31 17:49:10 +08:00
magic.chen
9236f4260e llm proxy framework design, adding support for bard large language model proxy (#376) 
1. llm proxy framework design, adding support for bard large language
model proxy
2. add Alibaba Cloud Image Deployment Solution.

Close #369
 2023-07-28 17:04:09 +08:00
xuyuan23
4da524340c refactor: rename the data privacy module 
rename data privace module.
 2023-07-28 14:45:27 +08:00
xuyuan23
ca50e87b8f fix: support bard in chatdb 
support bard proxy llm  in chatdb
 2023-07-27 19:22:23 +08:00
xuyuan23
c4638bfde3 fix: fix function bard_generate_stream's output 
fix function bard_generate_stream's output
 2023-07-27 11:11:59 +08:00
xuyuan23
572c336b81 feat: Adjusting the proxy framework. 
Adjusting the proxy framework, chatgpt_proxyllm default, or proxyllm instead, new LLM proxy should be configured in format 'xxx_proxyllm'.
 2023-07-27 11:03:29 +08:00
xuyuan23
b1d84fb317 fix: fix bard LLM request param error 
fix bard LLM request param error, use 'content' as the input
 2023-07-26 17:02:48 +08:00
xuyuan23
d95726c04a feat: add Bard LLM proxy 
Supplement proxy framework content, adding proxy access method for the Bard LLM.

Closes #369
 2023-07-26 16:36:26 +08:00
FangYin Cheng
4b1c0a36f8 feat: Support baichuan-7B model 2023-07-25 22:08:07 +08:00
FangYin Cheng
7baaaf0849 feat: Support baichuan-13B model 2023-07-25 22:07:02 +08:00
FangYin Cheng
35d0c17ae8 feat: Support baichuan-7B model 2023-07-25 19:48:21 +08:00
FangYin Cheng
01074660bc feat: Support baichuan-13B model 2023-07-25 19:48:21 +08:00
FangYin Cheng
168c754a3f feat: Support llama-2 model 2023-07-20 21:44:40 +08:00
FangYin Cheng
d805e6ab42 feat: structured prompt data interaction between dbgpt-server and llm-server 2023-07-19 12:10:22 +08:00