 oobabooga
|
7c4d5ca8cc
Improve the text generation call a bit
|
2 lat temu |
|
oobabooga
|
0f6708c471
Sort the imports
|
2 lat temu |
|
oobabooga
|
e735806c51
Add a generate() function for RWKV
|
2 lat temu |
|
oobabooga
|
f871971de1
Trying to get the chat to work
|
2 lat temu |
|
oobabooga
|
ebd698905c
Add streaming to RWKV
|
2 lat temu |
|
oobabooga
|
70e522732c
Move RWKV loader into a separate file
|
2 lat temu |
|
oobabooga
|
ebc64a408c
RWKV support prototype
|
2 lat temu |
|
oobabooga
|
6e843a11d6
Fix FlexGen in chat mode
|
2 lat temu |
|
oobabooga
|
fa58fd5559
Proper way to free the cuda cache
|
2 lat temu |
|
oobabooga
|
700311ce40
Empty the cuda cache at model.generate()
|
2 lat temu |
|
oobabooga
|
78ad55641b
Remove duplicate max_new_tokens parameter
|
2 lat temu |
|
oobabooga
|
65326b545a
Move all gradio elements to shared (so that extensions can use them)
|
2 lat temu |
|
oobabooga
|
9ae063e42b
Fix softprompts when deepspeed is active (#112)
|
2 lat temu |
|
oobabooga
|
7224343a70
Improve the imports
|
2 lat temu |
|
oobabooga
|
1dacd34165
Further refactor
|
2 lat temu |
|
oobabooga
|
ce7feb3641
Further refactor
|
2 lat temu |
