Upload folder using huggingface_hub

Dockerfile

@@ -27,8 +27,8 @@ RUN apt-get install git g++ python3 python3-pip -y && apt-get clean
 
 COPY pull06.sh pull06.sh
 RUN /bin/bash -x pull06.sh
-COPY pull8.sh pull8.sh
-RUN /bin/bash -x pull8.sh
+# COPY pull8.sh pull8.sh
+# RUN /bin/bash -x pull8.sh
 # COPY pull14.sh pull14.sh
 # RUN /bin/bash -x pull14.sh
 

main.py

@@ -1,5 +1,6 @@
 from flask import Flask
 import ollama
+from util import time_model
 
 app = Flask(__name__)
 
@@ -10,9 +11,9 @@ def hello_world():
 
 @app.route('/list')
 def test_world():
-    from ollama import ListResponse, list
+    from ollama import ListResponse, list  as ollist
 
-    response: ListResponse = list()
+    response: ListResponse = ollist()
     ansver = []
     for model in response.models:
       ansver.append(f'Name:{model.model}')
@@ -42,7 +43,7 @@ def ps():
           ansver.append(f'  Context length:  {model.context_length}')
           ansver.append(f'\n')
 
-    response: ListResponse = list()
+    response: ListResponse = ollist()
     for model in response.models:
       ansver.append(f'Name:{model.model}')
       ansver.append(f'  Size (MB): {(model.size.real / 1024 / 1024):.2f}')
@@ -84,85 +85,17 @@ def test_time():
 
 @app.route('/time8')
 def test_time8():
-  from  datetime import datetime
-  from ollama import Client
-  from ollama import chat
-
-  t_start=datetime.now()
-  ansver = []
-  messages = [
-  {
-    'role': 'user',
-    'content': 'Расскажи о себе подробно',
-  },
-  ]
-  response = chat('qwen3:8b', messages=messages)
-  ansver.append(f' start {t_start} ')
-  response_time=datetime.now()-t_start
-  ansver.append(f' duration  {response_time} ')
-  #
-  response_len = len(response['message']['content'])
-  ansver.append(f" lehgth {response_len}")
-  response_speed=response_len/int(response_time.seconds)
-  ansver.append(f" token/sek {response_len}")
-  ansver.append(response['message']['content'])
-  result = [f'<p>{answ}</p>' for answ in ansver ]
-  return '\n'.join(result)
+  return time_model('qwen3:8b')
 
 #@app.route('/time14')
 def test_time14():
-  from  datetime import datetime
-  from ollama import Client
-  from ollama import chat
-
-  t_start=datetime.now()
-  ansver = []
-  messages = [
-  {
-    'role': 'user',
-    'content': 'Расскажи о себе подробно',
-  },
-  ]
-  response = chat('qwen3:14b', messages=messages)
-  ansver.append(f' start {t_start} ')
-  response_time=datetime.now()-t_start
-  ansver.append(f' duration  {response_time} ')
-  #
-  response_len = len(response['message']['content'])
-  ansver.append(f" lehgth {response_len}")
-  response_speed=response_len/int(response_time.seconds)
-  ansver.append(f" token/sek {response_len}")
-  ansver.append(response['message']['content'])
-  result = [f'<p>{answ}</p>' for answ in ansver ]
-  return '\n'.join(result)
+  return time_model('qwen3:14b')
 
 
 #@app.route('/time30')
 def test_time30():
-  from  datetime import datetime
-  from ollama import Client
-  from ollama import chat
+  return time_model('qwen3:30b')
 
-  t_start=datetime.now()
-  ansver = []
-  messages = [
-  {
-    'role': 'user',
-    'content': 'Расскажи о себе подробно',
-  },
-  ]
-  response = chat('qwen3:30b', messages=messages)
-  ansver.append(f' start {t_start} ')
-  response_time=datetime.now()-t_start
-  ansver.append(f' duration  {response_time} ')
-  #
-  response_len = len(response['message']['content'])
-  ansver.append(f" lehgth {response_len}")
-  response_speed=response_len/int(response_time.seconds)
-  ansver.append(f" token/sek {response_len}")
-  ansver.append(response['message']['content'])
-  result = [f'<p>{answ}</p>' for answ in ansver ]
-  return '\n'.join(result)
 
 if __name__ == '__main__':
    app.run(host='0.0.0.0', port="7860")

start.sh

@@ -6,6 +6,7 @@ while true
   done >/dev/null 2>&1 & 
 echo  sleep 3
 sleep 3
+/bin/bash pull8.sh
 /bin/bash pull14.sh
 echo   list
 /usr/bin/ollama list

util.py

@@ -0,0 +1,168 @@
+from flask import Flask
+import ollama
+
+app = Flask(__name__)
+
+@app.route('/')
+def hello_world():
+    return "<h1>Hello, World!<h1>"
+
+
+@app.route('/list')
+def test_world():
+    from ollama import ListResponse, list
+
+    response: ListResponse = list()
+    ansver = []
+    for model in response.models:
+      ansver.append(f'Name:{model.model}')
+      ansver.append(f'  Size (MB): {(model.size.real / 1024 / 1024):.2f}')
+      if model.details:
+        ansver.append(f'  Format: {model.details.format}')
+        ansver.append(f'  Family:  {model.details.family}')
+        ansver.append(f'  Parameter Size:  {model.details.parameter_size}')
+        ansver.append(f'  Quantization Level:{model.details.quantization_level}')
+    result = [f'<p>{answ}</p>' for answ in ansver ]
+    return '\n'.join(result)
+
+
+@app.route('/ps')
+def ps():
+    from ollama import ProcessResponse, chat, ps, pull
+    ansver = []
+
+    response: ProcessResponse = ps()
+    for model in response.models:
+          ansver.append(f'Model:  {model.model}')
+          ansver.append(f'  Digest:  {model.digest}')
+          ansver.append(f'  Expires at: {model.expires_at}')
+          ansver.append(f'  Size:  {model.size}')
+          ansver.append(f'  Size vram:  {model.size_vram}')
+          ansver.append(f'  Details: {model.details}')
+          ansver.append(f'  Context length:  {model.context_length}')
+          ansver.append(f'\n')
+
+    response: ListResponse = list()
+    for model in response.models:
+      ansver.append(f'Name:{model.model}')
+      ansver.append(f'  Size (MB): {(model.size.real / 1024 / 1024):.2f}')
+      if model.details:
+        ansver.append(f'  Format: {model.details.format}')
+        ansver.append(f'  Family:  {model.details.family}')
+        ansver.append(f'  Parameter Size:  {model.details.parameter_size}')
+        ansver.append(f'  Quantization Level:{model.details.quantization_level}')
+    result = [f'<p>{answ}</p>' for answ in ansver ]
+    return '\n'.join(result)
+
+@app.route('/time')
+def test_time():
+  from  datetime import datetime
+  from ollama import Client
+  from ollama import chat
+
+  t_start=datetime.now()
+  ansver = []
+  messages = [
+  {
+    'role': 'user',
+    'content': 'Расскажи о себе подробно',
+  },
+  ]
+  response = chat('qwen3:0.6b', messages=messages)
+  ansver.append(f' start {t_start} ')
+  response_time=datetime.now()-t_start
+  ansver.append(f' duration  {response_time} ')
+  #
+  response_len = len(response['message']['content'])
+  ansver.append(f" lehgth {response_len}")
+  response_speed=response_len/int(response_time.seconds)
+  ansver.append(f" token/sek {response_len}")
+  ansver.append(response['message']['content'])
+  result = [f'<p>{answ}</p>' for answ in ansver ]
+  return '\n'.join(result)
+
+
+@app.route('/time8')
+def test_time8():
+  from  datetime import datetime
+  from ollama import Client
+  from ollama import chat
+
+  t_start=datetime.now()
+  ansver = []
+  messages = [
+  {
+    'role': 'user',
+    'content': 'Расскажи о себе подробно',
+  },
+  ]
+  response = chat('qwen3:8b', messages=messages)
+  ansver.append(f' start {t_start} ')
+  response_time=datetime.now()-t_start
+  ansver.append(f' duration  {response_time} ')
+  #
+  response_len = len(response['message']['content'])
+  ansver.append(f" lehgth {response_len}")
+  response_speed=response_len/int(response_time.seconds)
+  ansver.append(f" token/sek {response_len}")
+  ansver.append(response['message']['content'])
+  result = [f'<p>{answ}</p>' for answ in ansver ]
+  return '\n'.join(result)
+
+#@app.route('/time14')
+def test_time14():
+  from  datetime import datetime
+  from ollama import Client
+  from ollama import chat
+
+  t_start=datetime.now()
+  ansver = []
+  messages = [
+  {
+    'role': 'user',
+    'content': 'Расскажи о себе подробно',
+  },
+  ]
+  response = chat('qwen3:14b', messages=messages)
+  ansver.append(f' start {t_start} ')
+  response_time=datetime.now()-t_start
+  ansver.append(f' duration  {response_time} ')
+  #
+  response_len = len(response['message']['content'])
+  ansver.append(f" lehgth {response_len}")
+  response_speed=response_len/int(response_time.seconds)
+  ansver.append(f" token/sek {response_len}")
+  ansver.append(response['message']['content'])
+  result = [f'<p>{answ}</p>' for answ in ansver ]
+  return '\n'.join(result)
+
+
+#@app.route('/time30')
+def test_time30():
+  from  datetime import datetime
+  from ollama import Client
+  from ollama import chat
+
+  t_start=datetime.now()
+  ansver = []
+  messages = [
+  {
+    'role': 'user',
+    'content': 'Расскажи о себе подробно',
+  },
+  ]
+  response = chat('qwen3:30b', messages=messages)
+  ansver.append(f' start {t_start} ')
+  response_time=datetime.now()-t_start
+  ansver.append(f' duration  {response_time} ')
+  #
+  response_len = len(response['message']['content'])
+  ansver.append(f" lehgth {response_len}")
+  response_speed=response_len/int(response_time.seconds)
+  ansver.append(f" token/sek {response_len}")
+  ansver.append(response['message']['content'])
+  result = [f'<p>{answ}</p>' for answ in ansver ]
+  return '\n'.join(result)
+
+if __name__ == '__main__':
+   app.run(host='0.0.0.0', port="7860")