srijaydeshpande commited on
Commit
67878ad
·
verified ·
1 Parent(s): f3c0445

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +9 -3
app.py CHANGED
@@ -16,9 +16,15 @@ from llama_cpp import Llama
16
  # local_dir = "./models"
17
  # )
18
 
 
 
 
 
 
 
19
  hf_hub_download(
20
- repo_id="bartowski/Meta-Llama-3-70B-Instruct-GGUF",
21
- filename="Meta-Llama-3-70B-Instruct-Q3_K_M.gguf",
22
  local_dir = "./models"
23
  )
24
 
@@ -93,7 +99,7 @@ def pdf_to_text(cv_file, prompt, maxtokens=2048, temperature=0, top_probability=
93
  for page_id in page2content:
94
  cv_text += page2content[page_id] + ' '
95
  llm = Llama(
96
- model_path="models/Meta-Llama-3-70B-Instruct-Q3_K_M.gguf",
97
  flash_attn=True,
98
  n_gpu_layers=81,
99
  n_batch=1024,
 
16
  # local_dir = "./models"
17
  # )
18
 
19
+ # hf_hub_download(
20
+ # repo_id="bartowski/Meta-Llama-3-70B-Instruct-GGUF",
21
+ # filename="Meta-Llama-3-70B-Instruct-Q3_K_M.gguf",
22
+ # local_dir = "./models"
23
+ # )
24
+
25
  hf_hub_download(
26
+ repo_id="leafspark/Meta-Llama-3.1-405B-Instruct-GGUF",
27
+ filename="Llama-3.1-405B-Instruct.Q2_K.gguf",
28
  local_dir = "./models"
29
  )
30
 
 
99
  for page_id in page2content:
100
  cv_text += page2content[page_id] + ' '
101
  llm = Llama(
102
+ model_path="models/Llama-3.1-405B-Instruct.Q2_K.gguf",
103
  flash_attn=True,
104
  n_gpu_layers=81,
105
  n_batch=1024,