Lyte commited on
Commit
372a5eb
·
verified ·
1 Parent(s): a35d071

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +12 -10
app.py CHANGED
@@ -1,18 +1,21 @@
1
  import os
2
  import gradio as gr
3
  from llama_cpp import Llama
4
- from huggingface_hub import hf_hub_download
 
 
 
5
 
6
  model = Llama(
7
  model_path=hf_hub_download(
8
- repo_id=os.environ.get("REPO_ID", "bartowski/QwQ-32B-Preview-GGUF"),
9
- filename=os.environ.get("MODEL_FILE", "QwQ-32B-Preview-Q3_K_L.gguf"),
10
  )
11
  )
12
 
13
  DESCRIPTION = '''
14
- # QwQ-32B-Preview | Duplicate the space and set it to private for faster & personal inference for free.
15
- Qwen/QwQ-32B-Preview: an experimental research model developed by the Qwen Team.
16
  Focused on advancing AI reasoning capabilities.
17
 
18
  **To start a new chat**, click "clear" and start a new dialog.
@@ -26,8 +29,7 @@ def generate_text(message, history, max_tokens=512, temperature=0.9, top_p=0.95)
26
  """Generate a response using the Llama model."""
27
  temp = ""
28
  response = model.create_chat_completion(
29
- messages=[{"role": "system", "content": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."},
30
- {"role": "user", "content": message}],
31
  temperature=temperature,
32
  max_tokens=max_tokens,
33
  top_p=top_p,
@@ -44,12 +46,12 @@ with gr.Blocks() as demo:
44
 
45
  chatbot = gr.ChatInterface(
46
  generate_text,
47
- title="Qwen/QwQ-32B-Preview | GGUF Demo",
48
  description=" settings below if needed.",
49
  examples=[
50
  ["How many r's are in the word strawberry?"],
51
- ['What is the most optimal way to do Test-Time Scaling?'],
52
- ['Explain to me how gravity works like I am 5!'],
53
  ],
54
  cache_examples=False,
55
  fill_height=True
 
1
  import os
2
  import gradio as gr
3
  from llama_cpp import Llama
4
+ from huggingface_hub import hf_hub_download, login
5
+ import os
6
+
7
+ login(os.getenv("HF_TOKEN"))
8
 
9
  model = Llama(
10
  model_path=hf_hub_download(
11
+ repo_id=os.environ.get("REPO_ID", "Lyte/HuatuoGPT-o1-7B-Q4_K_M-GGUF"),
12
+ filename=os.environ.get("MODEL_FILE", "huatuogpt-o1-7b-q4_k_m.gguf"),
13
  )
14
  )
15
 
16
  DESCRIPTION = '''
17
+ # FreedomIntelligence/HuatuoGPT-o1-7B | Duplicate the space and set it to private for faster & personal inference for free.
18
+ HuatuoGPT-o1-7B: an experimental research model developed by the Qwen Team.
19
  Focused on advancing AI reasoning capabilities.
20
 
21
  **To start a new chat**, click "clear" and start a new dialog.
 
29
  """Generate a response using the Llama model."""
30
  temp = ""
31
  response = model.create_chat_completion(
32
+ messages=[{"role": "user", "content": message}],
 
33
  temperature=temperature,
34
  max_tokens=max_tokens,
35
  top_p=top_p,
 
46
 
47
  chatbot = gr.ChatInterface(
48
  generate_text,
49
+ title="FreedomIntelligence/HuatuoGPT-o1-7B | GGUF Demo",
50
  description=" settings below if needed.",
51
  examples=[
52
  ["How many r's are in the word strawberry?"],
53
+ ['How to stop a cough?'],
54
+ ['How do I relieve feet pain?'],
55
  ],
56
  cache_examples=False,
57
  fill_height=True