tanyuzhou commited on
Commit
c7a58b6
·
1 Parent(s): d4e8b9d
Files changed (2) hide show
  1. app.py +1 -1
  2. requirements.txt +1 -1
app.py CHANGED
@@ -7,7 +7,7 @@ from transformers import TextStreamer
7
  import spaces
8
 
9
  quantization_config = BitsAndBytesConfig(
10
- load_in_4bit=True, bnb_4bit_compute_dtype=torch.float16
11
  )
12
 
13
  # Load model and tokenizer
 
7
  import spaces
8
 
9
  quantization_config = BitsAndBytesConfig(
10
+ load_in_4bit=True, bnb_4bit_use_double_quant=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.bfloat16
11
  )
12
 
13
  # Load model and tokenizer
requirements.txt CHANGED
@@ -1,5 +1,5 @@
1
  accelerate
2
- bitsandbytes==0.43.0
3
  timm
4
  einops
5
  torch
 
1
  accelerate
2
+ bitsandbytes==0.42.0
3
  timm
4
  einops
5
  torch